system

Abstract

Provide a system. A method for collecting customer voice data; means for converting the voice data into text data; means for analyzing and classifying said text data; means for allocating the classified data to a corresponding department; means for generating proposals based on the assigned data; means for notifying the generated proposal content; A system including:

Description

[Technical Field]

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

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

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

[0004] There is a need to efficiently collect and analyze customer voice data and quickly and appropriately assign it to the appropriate department to improve responses to customer requests. However, current systems lack the accuracy of voice data conversion and analytical capabilities, making this process inefficient. Furthermore, because voice data cannot be properly classified, it is often not possible to assign it to the appropriate department, resulting in delayed responses. This leads to problems such as reduced customer satisfaction and delayed response times. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system including: means for collecting customer voice data; means for converting the voice data into text data; means for analyzing and classifying the text data; means for assigning the classified data to the corresponding department; means for generating proposals based on the assigned data; and means for notifying the generated proposals. Specifically, the system collects voice data in real time and converts it into text with high accuracy using a voice recognition engine. The system then analyzes the text data using natural language processing technology, extracting keywords, analyzing sentiment, and identifying topics, and classifying them into appropriate categories. This enables prompt and appropriate responses to each department, improving customer satisfaction and shortening response times.

[0006] "Customer voice data" is data in the form of voice collected from a customer.

[0007] "Means for collection" refers to a system or device for collecting customer voice data in real time or in batch processing.

[0008] "Means for converting voice data into text data" refers to a system or method that uses a voice recognition engine to convert voice data into text information.

[0009] A "means for analyzing and classifying text data" is a system or method that uses natural language processing technology to analyze text data and classify it to belong to a specific category.

[0010] "Means for allocating to corresponding departments" refers to a system or method for distributing classified data to the appropriate departments or sections within the company.

[0011] The "means for generating proposal content" refers to a system or method for automatically creating proposals and countermeasures based on the data assigned to the corresponding department.

[0012] The "means for notifying the content of the generated proposal" refers to a system or method for notifying the corresponding department or person in charge of the automatically generated proposal.

[0013] "Storage means" refers to a system or method for recording and storing collected voice data and converted text data in a database or the like.

[0014] "Natural language processing technology" is a computer technology for analyzing text data and performing tasks such as extracting keywords, analyzing sentiment, and identifying topics.

[0015] "Sentiment analysis" is the process of determining the sentiment of text data from its content.

[0016] "Topic identification" is the process of identifying the subject or theme from the content of text data. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] This invention is a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies the customer of proposals. A specific embodiment of this system will be described.

[0039] System Overview

[0040] This system is mainly composed of three components: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0041] Audio data collection

[0042] The terminal captures voice data from the customer in real time or in batch processing, and the captured voice data is sent to the server and stored in a database.

[0043] Converting audio data to text

[0044] The server sends the saved voice data to a voice recognition engine, which converts the voice data into text data and saves the text data back in the database.

[0045] Text data analysis

[0046] The server analyzes the text data using natural language processing (NLP) techniques, which specifically perform the following operations:

[0047] Keyword extraction: Using specific algorithms to extract key keywords from text data.

[0048] Sentiment analysis: Determine the sentiment (positive, negative, neutral) from the content of text data.

[0049] Topic identification: Identifying themes and topics based on text data.

[0050] Category sorting

[0051] The server analyzes the text data and classifies it into predefined categories, such as "technical support," "sales," and "complaints."

[0052] Allocation to products

[0053] The server assigns categorized text data to the appropriate department, for example, data categorized in the "Tech Support" category is sent to the Tech Support department.

[0054] Proposal / proposal

[0055] The server automatically generates proposals based on the assigned data. The generated proposals are notified to the person in charge (user) of the corresponding department via the notification system. The user then checks the notified proposals and takes specific action.

[0056] Specific examples

[0057] As a concrete example, consider the case of processing voice data from a customer saying, "The network connection is unstable." The device collects this voice data, and the server uses a speech recognition engine to convert it into text data saying, "The network connection is unstable." The server then analyzes this text data using NLP technology to extract the keywords "network," "connection," and "unstable," and determines that the sentiment is negative. This data is classified into the "technical support" category and assigned to the technical support department. Finally, the server generates a proposal, "technical support procedures for stabilizing the network connection," and notifies the technical support representative (user). The user then provides specific technical support to the customer based on this proposal.

[0058] In this way, this system efficiently processes voice data from customers and enables quick and appropriate responses, thereby improving customer satisfaction and operational efficiency.

[0059] The processing flow will be explained below.

[0060] Step 1: Collecting audio data

[0061] The terminal receives calls and voice messages from customers. The received voice data is sent to the server in real time or in batch processing. The server stores this voice data in a database.

[0062] Step 2: Convert audio data to text

[0063] The server sends the voice data stored in the database to a voice recognition engine, which converts the voice data into text data, which is then stored in the database by the server.

[0064] Step 3: Analyze the text data

[0065] The server analyzes the text data using natural language processing (NLP) techniques. The server applies keyword extraction algorithms to extract important keywords. Furthermore, the server performs sentiment analysis to determine the sentiment in the text (e.g., positive, negative, neutral). Finally, the server performs topic identification to recognize the subject matter of the text data.

[0066] Step 4: Categorize

[0067] Based on the analysis, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[0068] Step 5: Allocate to products

[0069] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[0070] Step 6: Generate proposals

[0071] The server automatically generates suggestions based on the assigned text data. For example, if the data is related to technical support, it generates technical procedures for stabilizing network connections.

[0072] Step 7: Notification of proposal

[0073] The server sends the generated proposal to the notification system, which then notifies the relevant department (user) of the proposal. The user then checks the proposal and takes specific action.

[0074] The above processing steps enable accurate understanding of customer feedback and prompt and appropriate response.

[0075] Example 1

[0076] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0077] Conventional systems have difficulty efficiently collecting voice data from customers, appropriately converting it into text data, analyzing it, classifying it, and generating and notifying customers of proposals. In particular, the lack of keyword extraction, sentiment analysis, or topic identification during voice data analysis means that appropriate responses tend to be delayed. Furthermore, because proposals are not automatically generated using generative AI models, there is room for improvement in the accuracy and speed of responses. There is a need for a system that can resolve these issues and enable faster and more accurate responses.

[0078] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0079] In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data by extracting keywords, analyzing sentiment, and identifying topics using natural language processing technology, means for assigning the classified data to an appropriate department based on predefined categories, means for generating proposals using a generative AI model based on the assigned data, and means for notifying the generated proposals. This enables efficient and accurate processing of voice data from customers, enabling prompt and accurate responses.

[0080] "Customer voice data" refers to voice information transmitted by customers, including customer questions, complaints, requests, etc.

[0081] "Means for collection" refers to the device or software used to capture voice data from customers and input it into the system.

[0082] "Means for converting voice data into text data" refers to a device or software that converts voice data into text data using voice recognition technology.

[0083] "Means for analyzing text data" refers to a device or software that uses natural language processing techniques to analyze acquired text data.

[0084] "Means for classifying" refers to a device or software that appropriately classifies analyzed text data into predefined categories.

[0085] A "predefined category" is a category that is set in advance for classifying data, and examples include "technical support," "sales," and "complaints."

[0086] "Means for allocating to appropriate departments" refers to devices or software that transmit or distribute classified data to the departments corresponding to each category.

[0087] "Generative AI model" refers to an artificial intelligence model that uses deep learning technology to generate suggestions based on specific input data.

[0088] "Means for generating proposals" refers to devices or software that use generative AI models to automatically create proposals and solutions based on customer feedback.

[0089] "Means for notifying" refers to a device or software for communicating the generated proposal to an appropriate recipient.

[0090] A "prompt sentence" refers to text data that is input to a generative AI model to generate a specific output.

[0091] The present invention is a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies the customer of proposals. Specific embodiments are described below.

[0092] This system is mainly composed of three entities: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0093] Audio data collection

[0094] The terminal acquires voice data from the customer through a user-operated interface. Specifically, the terminal is equipped with a microphone and recording software (e.g., Audacity) that is used to collect voice data in real time or in batches. The collected voice data is encrypted and transmitted to a server using a secure communication protocol (e.g., HTTPS).

[0095] Converting audio data to text

[0096] The server stores the received voice data in a database. The server then sends the stored voice data to a speech recognition engine (e.g., Google® Cloud Speech-to-Text API) to convert the voice data into text data, which is then stored in the database again.

[0097] Text data analysis

[0098] The server analyzes the stored text data using natural language processing (NLP) techniques, specifically performing the following operations:

[0099] Keyword extraction: Uses the TF-IDF algorithm to extract key keywords from text data.

[0100] Sentiment analysis: Use a sentiment analysis API (e.g., IBM Watson® Natural Language Understanding) to classify the sentiment of text data as positive, negative, or neutral.

[0101] Topic Identification: Identify topics within text data using pre-trained topic models (e.g., Latent Dirichlet Allocation).

[0102] Category sorting

[0103] Based on the results of the text data analysis, the server classifies the data into predefined categories (e.g., "technical support," "sales," "complaints") using the Naive Bayes classifier as the classification algorithm.

[0104] Allocation to products

[0105] The server assigns the classified text data to the appropriate department, using a department collaboration system (e.g., Slack or MICROSOFT® TEAMS®).

[0106] Proposal / proposal

[0107] The server automatically generates proposals using a generative AI model (e.g., OpenAI's GPT-3) based on the assigned data. The generated proposals are notified to the responsible party via a notification system (e.g., email or SMS gateway). The user receives the notification and takes specific action.

[0108] Specific examples

[0109] As a specific example, consider a case where voice data from a customer saying "The network connection is unstable" is collected by a terminal.

[0110] 1. The device collects the customer's voice message saying "The network connection is unstable," encrypts it, and sends it to the server.

[0111] 2. The server stores the voice data in a database and uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert it into text data such as "The network connection is unstable."

[0112] 3. The server analyzes this text data using natural language processing technology (e.g., IBM Watson Natural Language Understanding), extracts the keywords "network," "connection," and "unstable," and determines that the sentiment is negative.

[0113] 4. This data is classified under the category "Technical Support" and assigned to the Technical Support department.

[0114] 5. The server generates a suggestion by inputting the following prompt into the generative AI model (e.g., OpenAI's GPT-3): "The network connection is unstable. Please suggest technical support steps to resolve this."

[0115] 6. The generated proposal, "Technical support procedure for stabilizing network connections," is notified to the technical support staff. The technical support staff (user) provides specific technical support to the customer based on this proposal.

[0116] In this way, the system efficiently processes voice data from customers, enabling quick and accurate responses.

[0117] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0118] Step 1: Collecting audio data

[0119] Input: Customer voice input

[0120] Output: Audio data saved on the device (file format: .wav, etc.)

[0121] Specific operation:

[0122] The terminal utilizes a microphone and recording software (e.g., Audacity) to record audio in real time through the customer interface.

[0123] A customer says, "My network connection is unstable."

[0124] The device captures the audio and temporarily stores it in .wav format.

[0125] The saved audio data is encrypted and sent to the server.

[0126] Step 2: Convert audio data to text

[0127] Input: Collected encrypted audio data

[0128] Output: Text data

[0129] Specific operation:

[0130] The server stores the received voice data in a database.

[0131] The saved voice data is sent to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) and converted into text data.

[0132] The speech recognition engine returns text data saying "Network connection is unstable."

[0133] The server stores the converted text data in a database.

[0134] Step 3: Analyze the text data

[0135] Input: Text data

[0136] Output: Analysis results (keywords, sentiment, topics)

[0137] Specific operation:

[0138] The server analyzes the stored text data using natural language processing (NLP) technology.

[0139] Keyword extraction: The keywords “network”, “connection”, and “unstable” are extracted using the TF-IDF algorithm.

[0140] Sentiment analysis: Use a sentiment analysis API (e.g., IBM Watson Natural Language Understanding) to determine whether the text data is negative.

[0141] Topic Identification: Identify the topic "stabilizing network connectivity" using a pre-trained topic model (e.g., Latent Dirichlet Allocation).

[0142] Step 4: Categorize

[0143] Input: Analysis results (keywords, sentiment, topics)

[0144] Output: Classified category information

[0145] Specific operation:

[0146] Based on the analysis, the server classifies the data into predefined categories (e.g., "technical support").

[0147] The Naive Bayes classifier is used as the classification algorithm.

[0148] Information categorized under "Technical Support" is generated.

[0149] Step 5: Assign to departments

[0150] Input: Classified category information

[0151] Output: Data assigned to departments

[0152] Specific operation:

[0153] The server assigns data categorized under the "Technical Support" category to the Technical Support department.

[0154] Send data to the technical support department in real time using a cross-departmental system (e.g., Slack or Microsoft Teams).

[0155] Step 6: Proposal and project creation

[0156] Input: Data assigned to department

[0157] Output: Generated proposals

[0158] Specific operation:

[0159] The server generates suggestions using a generative AI model (e.g., OpenAI's GPT-3) based on the assigned data.

[0160] Enter the prompt sentence into the generative AI model: "The network connection is unstable. Please suggest technical support steps to resolve it."

[0161] The generative AI model generates "technical support procedures for stabilizing network connections," which are output as suggestions.

[0162] Step 7: Notification

[0163] Input: Generated proposal

[0164] Output: The suggestion sent to the user

[0165] Specific operation:

[0166] The server notifies the generated proposal to the technical support personnel via a notification system (e.g., email or SMS gateway).

[0167] The user receives a notification and takes specific action.

[0168] (Application example 1)

[0169] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0170] Current food delivery systems lack the means to quickly and accurately collect and analyze customer feedback. In particular, there is no system that can efficiently process voice feedback, which results in issues such as time spent improving customer satisfaction and resolving problems. Furthermore, conventional systems have difficulty accurately understanding customer sentiment and the content of feedback and quickly directing the appropriate department to respond.

[0171] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0172] In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data, means for assigning the classified data to a corresponding department, means for generating proposals based on the assigned data, means for notifying the generated proposals, means for extracting keywords from customer feedback using the voice data, and means for performing sentiment analysis using natural language processing technology to classify categories based on the keywords. This makes it possible to efficiently collect and analyze customer voice feedback and quickly and appropriately notify the corresponding department of proposals.

[0173] "Customer voice data" refers to voice information generated by customers, including feedback and inquiries about services and products.

[0174] "Means" refers to devices, methods, techniques or processes used to achieve a particular purpose.

[0175] "Text data" refers to information converted from audio into a string of characters, such as a collection of sentences or words.

[0176] "Keyword extraction" refers to the process of selecting key words and phrases from text data.

[0177] "Sentiment analysis" is the process of determining the writer's emotions from the content of text data, which are classified as positive, negative, neutral, etc.

[0178] "Natural language processing technology" refers to the technology that enables computers to understand, interpret, and generate human language, and includes keyword extraction and sentiment analysis as part of this.

[0179] A "department" refers to a specialized team responsible for a specific task or function within an organization.

[0180] "Proposals" refer to solutions or countermeasures that are automatically generated based on the analyzed data.

[0181] "Real-time" refers to data collection and processing occurring immediately, with minimal delay.

[0182] "Sentiment analysis using natural language processing technology" refers to the technology of using a computer to analyze emotions from text data, and is a part of natural language processing technology.

[0183] A "system" refers to an overall structure in which multiple components and means work together to perform a specific function.

[0184] This invention provides a system for food delivery services that efficiently collects and analyzes customer voice feedback and promptly notifies the relevant department of the proposed solution. This system consists of three entities: a server, a terminal, and a user.

[0185] System Overview

[0186] server

[0187] The server handles the central data processing: converting voice data to text, analyzing text data, allocating data to products, generating recommendations, and notifications.

[0188] Terminal

[0189] The device collects voice data from customers in real time. The data is picked up through a microphone and sent to a server. The device also serves as an interface for customers and guides them through the feedback collection process.

[0190] User

[0191] The user checks the proposals sent by the system and implements specific measures. The user refers to the person in charge of the relevant department, such as a technical support staff member or a manager.

[0192] Audio data collection

[0193] The device collects voice data from the microphone using the speech_recognition library. The voice information spoken by the customer is captured in real time and sent to the server. The collected voice data is converted to text data using the recognizer.recognize_google method.

[0194] Text data analysis

[0195] The server analyzes the acquired text data using natural language processing technology, specifically by performing the following operations:

[0196] Keyword Extraction: Use TextBlob to extract key words and phrases from text data.

[0197] Sentiment Analysis: Use TextBlob to classify the sentiment of text (positive, negative, neutral).

[0198] Category classification: Uses the transformers library pipeline to determine which category the text belongs to.

[0199] Product allocation and proposal generation

[0200] Based on the analysis results, the server assigns the data to the appropriate department and generates a recommendation for the corresponding department. For example, if the request is "feedback on an order," it will be assigned to the order processing department and a recommendation such as "Please check the order."

[0201] Notification of proposal details

[0202] The generated proposals are automatically notified to the relevant departments. The proposals are provided instantly through a notification system (pseudo implementation).

[0203] Specific examples

[0204] Below is an example of customer feedback:

[0205] "I couldn't specify a delivery time."

[0206] "The food tastes different than usual"

[0207] "The app is hard to understand"

[0208] For these specific examples, gather audio feedback using the following example prompts:

[0209] Example prompt sentence:

[0210] "Please provide your feedback by voice. For example, 'I can't confirm my order!' or 'The food tastes different.'"

[0211] Hardware and software used

[0212] Hardware: Microphone (for audio input)

[0213] Software: Python, speech_recognition, transformers, TextBlob

[0214] Data processing and calculation

[0215] The server converts the voice data into text and performs keyword extraction, sentiment analysis, and categorization, allowing it to properly categorize customer feedback and quickly notify the relevant department with suggestions.

[0216] In this way, the present invention provides a system that improves customer satisfaction in food delivery services and also improves operational efficiency.

[0217] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0218] Step 1:

[0219] The terminal collects voice data. Specifically, it obtains voice feedback from the customer through a microphone and collects it in real time using a speech recognition engine (speech_recognition). The input is the customer's voice, and the output is voice data.

[0220] Step 2:

[0221] The terminal transmits the collected voice data to the server. The input is the voice data, and the output is the voice data transmitted to the server.

[0222] Step 3:

[0223] The server converts the received voice data into text data. Specifically, it converts the voice data into text using the recognizer.recognize_google method. The input is voice data, and the output is text data.

[0224] Step 4:

[0225] The server analyzes the converted text data. Specifically, it extracts keywords from the text data (using TextBlob), performs sentiment analysis (also using TextBlob), and classifies it into appropriate categories (using a pipeline from the transformers library). The input is text data, and the output is keywords, sentiment classification, and categorized data.

[0226] Step 5:

[0227] The server assigns data to the corresponding department based on the analysis results. The input is keywords, sentiment classification, and categorized data, and the output is the assignment of data to the corresponding department.

[0228] Step 6:

[0229] The server generates proposals based on the assigned data. Specifically, it automatically proposes predefined countermeasures based on the analyzed categories. The input is the data assigned to the corresponding department, and the output is the proposals.

[0230] Step 7:

[0231] The server notifies the person in charge of the corresponding department of the content of the generated proposal via the notification system. The input is the content of the proposal, and the output is the notification to the person in charge.

[0232] Step 8:

[0233] The user (person in charge) implements specific measures based on the proposal. The input is the proposal, and the output is the implementation of specific measures.

[0234] This series of steps enables efficient processing of voice feedback from customers and provides prompt and appropriate responses.

[0235] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0236] The present invention relates to a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies proposals. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to respond more accurately. Specific embodiments of this system will be described below.

[0237] System Overview

[0238] This system consists of three components: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0239] Audio data collection

[0240] The terminal captures voice data from the customer in real time or in batch processing, and the captured voice data is sent to the server and stored in a database.

[0241] Converting audio data to text

[0242] The server sends the saved voice data to a voice recognition engine, which converts the voice data into text data, which is then saved in a database by the server.

[0243] Text data analysis

[0244] The server analyzes the text data using natural language processing (NLP) techniques and an emotion engine. Specifically, this process involves the following operations:

[0245] Keyword extraction: Using specific algorithms to extract key keywords from text data.

[0246] Sentiment analysis: Using a sentiment engine to determine user sentiment from the content of text data (e.g., positive, negative, neutral).

[0247] Topic identification: Identifying themes and topics based on text data.

[0248] Category sorting

[0249] Based on the analysis results, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[0250] Allocation to products

[0251] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[0252] Proposal / proposal

[0253] The server automatically generates suggestions based on the assigned text data. The generated suggestions reflect the emotions determined by the emotion engine. For example, if negative emotions are included, the server will suggest response procedures that require special attention.

[0254] Notification of proposal details

[0255] The server sends the generated proposal to the notification system, which then notifies the relevant department (user) of the proposal. The user then checks the proposal and takes specific action.

[0256] Specific examples

[0257] As a concrete example, consider the case of processing voice data from a customer saying, "The network connection is unstable." The device collects this voice data, and the server uses a speech recognition engine to convert it into text data saying, "The network connection is unstable." The server then analyzes this text data using NLP technology and an emotion engine to extract the keywords "network," "connection," and "unstable," and determines that the emotion is negative. This data is classified into the "technical support" category and assigned to the technical support department. Finally, the server generates a proposal saying, "Technical support procedures for stabilizing the network connection, and particularly, a prompt response is required," and notifies the technical support staff (user). The user then provides specific technical support to the customer based on this proposal.

[0258] In this way, this system efficiently processes voice data from customers and, by combining it with an emotion engine, provides more accurate responses, enabling quick and appropriate responses.

[0259] The processing flow will be explained below.

[0260] Step 1: Collecting audio data

[0261] The terminal receives calls and voice messages from customers in real time, and the received voice data is sent from the terminal to a server, which stores the voice data in a database.

[0262] Step 2: Convert audio data to text

[0263] The server sends the voice data stored in the database to a voice recognition engine, which converts the voice data into text data, which is then stored in the database by the server.

[0264] Step 3: Analyze the text data

[0265] The server analyzes the text data using natural language processing (NLP) technology and an emotion engine. Specifically, it uses NLP technology to perform keyword extraction and identify important keywords. It then uses the emotion engine to determine the sentiment (positive, negative, neutral) in the text data. It also performs topic identification and identifies the main themes in the text data.

[0266] Step 4: Categorize

[0267] Based on the analysis, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[0268] Step 5: Allocate to products

[0269] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[0270] Step 6: Generate proposals

[0271] The server automatically generates suggestions based on the assigned text data. The suggestions reflect the emotions determined by the emotion engine. For example, if negative emotions are included, the server will suggest procedures that require special attention.

[0272] Step 7: Notification of proposal

[0273] The server sends the generated proposal to the notification system, which notifies the person in charge (user) of the proposal.

[0274] Step 8: User Action

[0275] The user receives the proposal from the notification system and takes specific action based on the proposal. For example, in the case of technical support, the user provides technical support to the customer based on the proposal.

[0276] The above processing steps enable the system to accurately understand customer feedback and respond quickly and appropriately. By combining this system with an emotion engine, the quality of responses can be further improved.

[0277] Example 2

[0278] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0279] Conventional customer support systems had problems in efficiently collecting and analyzing voice data from customers, allocating it to the appropriate department based on its content, and quickly generating and notifying proposals. In particular, they lacked the ability to respond in a way that took customer emotions into consideration, making it difficult to improve customer satisfaction.

[0280] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0281] In this invention, the server includes a means for the terminal to collect voice data in real time or batch processing and send it to the server, a means for the server to convert the voice data into text data using a voice recognition engine, and a means for the server to analyze the text data using natural language processing technology and an emotion engine. This makes it possible to efficiently collect and analyze voice data from customers, assign it to the appropriate department based on its content, and quickly generate and notify proposals. It also enables responses that take customer emotions into consideration, which is expected to improve customer satisfaction.

[0282] "Customer Voice Data" refers to information in voice format transmitted by a Customer.

[0283] "Collection means" refers to a combination of hardware and software for capturing and recording audio data.

[0284] "Means for converting into text data" refers to technology and devices that analyze voice data and convert it into corresponding character string information.

[0285] "Means of analysis and classification" refers to the process of processing text data, analyzing its content, and classifying it into specific categories.

[0286] "Means for allocating to the corresponding department" refers to a system or algorithm for transferring classified data to the most appropriate department.

[0287] "Means for generating proposal content" refers to technology that automatically creates countermeasures and proposal content based on analyzed data.

[0288] "Means for notifying" refers to communication devices and systems for notifying the relevant departments and personnel of the content of the generated proposal.

[0289] "Real-time or batch processing" refers to a method of processing data immediately and a method of processing data in bulk.

[0290] "Speech recognition engine" refers to a software and hardware system for converting voice data into text data.

[0291] "Natural language processing technology" refers to computer technologies and algorithms for understanding, analyzing, and generating text data.

[0292] An "emotion engine" refers to software and algorithms that analyze text data and determine emotions from it.

[0293] MODE FOR CARRYING OUT THE INVENTION

[0294] This invention relates to a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies proposals. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to respond more accurately. Specific embodiments are described below.

[0295] This system consists of three components: a server, a terminal, and a user. The server processes the main data, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0296] Audio data collection

[0297] The terminal captures voice data from the customer in real time or in batch processing using a microphone device and recording software (e.g., Audacity). The captured voice data is stored in temporary storage and later transmitted to the server.

[0298] Converting audio data to text

[0299] The server converts the received voice data into text data using a speech recognition engine (for example, Google Cloud Speech-to-Text API), and the converted text data is stored in a database.

[0300] Text data analysis

[0301] The server analyzes the text data using natural language processing technology and an emotion engine (e.g., Hume AI's emotion recognition API). The server extracts keywords using the TF-IDF algorithm and determines the user's emotion (positive, negative, neutral, etc.) using the emotion engine. It also uses natural language processing technology to identify topics.

[0302] Category sorting

[0303] The server then classifies the text data into predefined categories based on the analysis results (e.g., "technical support," "sales," "complaint," etc.) using the text classification capabilities of the Google Cloud Natural Language API.

[0304] Allocation to departments

[0305] The server assigns text data that has been classified into the appropriate category to the corresponding department, for example, data classified as "technical support" will be sent to the technical support department.

[0306] Proposal / proposal

[0307] The server automatically generates suggestions based on the assigned text data. The suggestions reflect the emotions determined by the emotion engine, and if negative emotions are included, a quick response is required.

[0308] Notification of proposal details

[0309] The server sends the generated proposal to the notification system. The notification system notifies the person in charge (user) of the proposal. The user confirms the notified proposal and takes specific action.

[0310] Specific examples

[0311] For example, consider the case where voice data from a customer saying "The network connection is unstable" is processed.

[0312] 1. The device collects and records this audio data.

[0313] 2. The server sends the audio file to the Google Cloud Speech-to-Text API and receives the text "Network connection is unstable."

[0314] 3. The server analyzes this text data using NLP technology and an emotion engine, extracts the keywords "network," "connection," and "unstable," and determines that the emotion is negative.

[0315] 4. The server categorizes this text data into the "Technical Support" category.

[0316] 5. The server sends this data to the technical support department.

[0317] 6. The server generates a suggestion: "Technical support procedures for stabilizing network connectivity and particularly rapid response is required."

[0318] 7. The server sends the generated proposal to the notification system, which then notifies the technical support staff (user). The user then provides specific technical support to the customer based on the proposal.

[0319] In this way, the system efficiently processes voice data from customers and, by combining it with an emotion engine, provides more accurate responses, enabling quick and appropriate responses.

[0320] Prompt Sentence Examples

[0321] For example, a prompt to a generative AI model might look like this:

[0322] "Convert the customer's voice data, 'The network connection is unstable,' into text data using NLP technology and a sentiment analysis engine, extract keywords, determine sentiment, and convert it into specific suggestions."

[0323] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0324] Step 1:

[0325] The terminal collects voice data in real time or in batch processing. It uses a microphone device and recording software (e.g., Audacity) to capture voice data from customers. This saves the customer's voice as an audio file, which is then sent to the server.

[0326] Specific operation: When a customer says "The network connection is unstable," the device's microphone records this, saves it as an audio file, and sends it to the server.

[0327] Input: Customer voice data

[0328] Output: Audio file

[0329] Step 2:

[0330] The server sends the received audio file to a speech recognition engine, which uses the Google Cloud Speech-to-Text API to convert the audio data into text data, which the server then stores in a database.

[0331] Specific operation: The server sends the audio file "The network connection is unstable" to the Google Cloud Speech-to-Text API, obtains the text data "The network connection is unstable", and stores it in the database.

[0332] Input: Audio file

[0333] Output: Text data

[0334] Step 3:

[0335] The server analyzes the stored text data using natural language processing technology and an emotion engine (for example, Hume AI's emotion recognition API). Specifically, it uses the TF-IDF algorithm to extract keywords, and the emotion engine determines the user's emotion (positive, negative, neutral, etc.). It also uses natural language processing technology to identify topics.

[0336] Specific operation: The server analyzes the text data "Your network connection is unstable," extracts the keywords "network," "connection," and "unstable," and determines that the content contains negative sentiment.

[0337] Input: Text data

[0338] Output: Keywords, sentiment analysis, identified topics

[0339] Step 4:

[0340] The server classifies the text data into predefined categories based on the analysis results, such as "technical support," "sales," and "complaints," using the Google Cloud Natural Language API.

[0341] Specific operation: The server classifies the text data "Network connection is unstable" into the "Technical Support" category.

[0342] Input: Keywords, sentiment analysis results, identified topics

[0343] Output: Categorized data

[0344] Step 5:

[0345] The server assigns text data that has been classified into the appropriate category to the corresponding department, for example, data classified as "technical support" is sent to the technical support department.

[0346] What happens: The server sends data categorized under "Tech Support" to the technical support department.

[0347] Input: Categorical data

[0348] Output: Data assigned to departments

[0349] Step 6:

[0350] The server automatically generates suggestions based on the assigned text data, taking into account the emotions (e.g., negative) determined by the emotion engine.

[0351] What happens: The server generates a suggestion that states, "Technical support procedures for stabilizing network connectivity, and particularly prompt action, are needed."

[0352] Input: Data assigned to department, emotion judgment result

[0353] Output: Proposal

[0354] Step 7:

[0355] The server sends the generated proposal to the notification system. The notification system notifies the person in charge (user) of the proposal. The user confirms the notified proposal and takes specific action.

[0356] Specific operation: The server sends the proposal to the notification system, which notifies the technical support staff (user). The user provides specific technical support based on the proposal.

[0357] Input: Proposal

[0358] Output: Proposal notified

[0359] (Application example 2)

[0360] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0361] Conventional customer service systems have had the problem of being unable to efficiently collect customer feedback, quickly analyze and classify it, and make appropriate proposals. In particular, when dealing with customers in physical stores, real-time feedback is required, but conventional systems were inadequate in this regard.

[0362] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data, means for assigning the classified data to the corresponding department, means for generating proposals based on the assigned data, means for notifying the generated proposals, and smartphone application means for responding to customers in real time at the store. This makes it possible to collect and analyze customer feedback in real time at a physical store and quickly make appropriate proposals.

[0363] "Customer voice data" refers to voice information of words and phrases spoken by a customer.

[0364] "Means of collection" refers to devices or systems for capturing audio data, such as microphones or voice recording applications.

[0365] "Means for converting into text data" refers to a voice recognition engine or software for converting voice data into text information.

[0366] "Means for analyzing and classifying" refers to natural language processing techniques and algorithms used to analyze the converted text data and classify it into specific categories.

[0367] The "means for allocating to the corresponding department" refers to a system, such as a database management system, that automatically sorts the analyzed and classified data to the appropriate business department.

[0368] "Means for generating recommendations" refers to software algorithms that automatically generate solutions or advice based on the assigned data.

[0369] "Means of notification" refers to messaging and alert systems that notify relevant department staff of generated suggestions and countermeasures.

[0370] "Customer service in stores" refers to the business activities of responding quickly to customer requests and questions in physical stores.

[0371] "Smartphone application means" refers to dedicated application software that allows users to respond to customers via their smartphones.

[0372] The system for implementing this invention is composed of three entities: a server, a terminal, and a user. The specific roles and processing contents of each entity are explained below.

[0373] Audio data collection and text conversion

[0374] The terminal uses a dedicated application installed on a smartphone to handle customer inquiries in real time within the store. The smartphone's microphone is used to collect customer voice data. This voice data is then sent to a server, where it is converted into text data using a speech recognition engine (for example, Google's speech recognition API). The converted text data is then stored on the server.

[0375] Analysis and classification of text data

[0376] The server then analyzes the converted text data using natural language processing (NLP) techniques, such as Hugging Face's BERT model, to extract keywords, analyze sentiment, and identify topics from the text data. It also uses an emotion engine to determine the user's sentiment (e.g., positive, negative, neutral) based on the content of the text data. Based on the analysis results, the text data is classified into predefined categories (e.g., technical support, sales, complaints).

[0377] Allocating data and generating proposals

[0378] The server automatically assigns the classified text data to the appropriate department. For example, data classified as "technical support" is sent to the technical support department. Next, a suggestion is automatically generated. The generated suggestion reflects the emotion determined by the emotion engine. If a negative emotion is included, a response procedure requiring special attention is suggested.

[0379] Notification and implementation of proposals

[0380] The server sends the generated proposal to a notification system, which notifies staff in the relevant departments. Notification methods include push notifications via smartphone applications and internal messaging systems. The users (staff) then check the notified proposal and implement specific measures. This enables quick and appropriate responses that reflect customer feedback in real time.

[0381] Adding specific examples

[0382] As a concrete example, consider the case where a customer in a store says, "I don't know where the product is." This voice data is collected by a smartphone app and converted into text data, "I don't know where the product is," using a voice recognition engine. The server then analyzes this text data using NLP technology and an emotion engine, extracting the keywords "product," "location," and "don't know," and determining that the emotion is negative. This data is classified into the "store support" category and assigned to the relevant department. Finally, the server generates a suggestion, "You need guidance to the product location and prompt customer service," and notifies the staff.

[0383] Prompt Sentence Examples

[0384] Voice data is collected and converted into text data. The converted text data is analyzed using NLP technology to extract keywords, analyze sentiment, and identify topics. Based on the results, appropriate suggestions are generated and notified to the relevant departments.

[0385] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0386] Step 1: Collecting audio data

[0387] The device (smartphone) collects the customer's voice data through the application using a microphone. The input is the customer's voice data, and the output is the collected voice file. This voice file is sent directly to the server.

[0388] Step 2: Convert audio data to text

[0389] The server receives the transmitted audio file and converts it into text data using a speech recognition engine. The input is the audio file, and the output is the converted text data. Specifically, the server analyzes the audio data using Google's speech recognition API and generates the corresponding text data.

[0390] Step 3: Save the text data

[0391] The server stores the converted text data in a database. The input of this process is the text data obtained in the previous step, and the output is the stored text data. A database management system is used to efficiently store the text data.

[0392] Step 4: Analyze and classify text data

[0393] The server analyzes the stored text data using natural language processing (NLP) technology and a sentiment analysis engine. Specifically, it uses the Hugging Face BERT model to extract keywords, analyze sentiment, and identify topics from the text data. The input is the text data, and the output is the extracted keywords, sentiment classification, and topic identification results.

[0394] Step 5: Allocate data

[0395] The server automatically assigns data to the appropriate department based on the analyzed and classified data. The input is keywords, sentiment classification, and topic identification results, and the output is department assignment information. This is done based on pre-defined categories.

[0396] Step 6: Generate proposals

[0397] The server generates appropriate proposals based on the assigned data. The input is the assignment information to the corresponding department, and the output is the generated proposal. The proposals are automatically created using an algorithm and include appropriate countermeasures.

[0398] Step 7: Notification of proposal

[0399] The server sends the generated proposal to a notification system, which notifies staff in the relevant departments. The input is the proposal, and the output is the notified information. Notification methods include push notifications on smartphones and internal messaging systems.

[0400] Step 8: Implementing customer interactions

[0401] The user (staff member) checks the notified proposal and takes specific action. The input is the notified proposal, and the output is the result of the action taken. This enables quick and appropriate action that reflects customer feedback in real time.

[0402] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0403] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0404] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0405] [Second embodiment]

[0406] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0407] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0408] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0409] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0410] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0411] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0412] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0413] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0414] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0415] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0416] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0417] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0418] This invention is a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies the customer of proposals. A specific embodiment of this system will be described.

[0419] System Overview

[0420] This system is mainly composed of three components: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0421] Audio data collection

[0422] The terminal captures voice data from the customer in real time or in batch processing, and the captured voice data is sent to the server and stored in a database.

[0423] Converting audio data to text

[0424] The server sends the saved voice data to a voice recognition engine, which converts the voice data into text data and saves the text data back in the database.

[0425] Text data analysis

[0426] The server analyzes the text data using natural language processing (NLP) techniques, which specifically perform the following operations:

[0427] Keyword extraction: Using specific algorithms to extract key keywords from text data.

[0428] Sentiment analysis: Determine the sentiment (positive, negative, neutral) from the content of text data.

[0429] Topic identification: Identifying themes and topics based on text data.

[0430] Category sorting

[0431] The server analyzes the text data and classifies it into predefined categories, such as "technical support," "sales," and "complaints."

[0432] Allocation to products

[0433] The server assigns categorized text data to the appropriate department, for example, data categorized in the "Tech Support" category is sent to the Tech Support department.

[0434] Proposal / proposal

[0435] The server automatically generates proposals based on the assigned data. The generated proposals are notified to the person in charge (user) of the corresponding department via the notification system. The user then checks the notified proposals and takes specific action.

[0436] Specific examples

[0437] As a concrete example, consider the case of processing voice data from a customer saying, "The network connection is unstable." The device collects this voice data, and the server uses a speech recognition engine to convert it into text data saying, "The network connection is unstable." The server then analyzes this text data using NLP technology to extract the keywords "network," "connection," and "unstable," and determines that the sentiment is negative. This data is classified into the "technical support" category and assigned to the technical support department. Finally, the server generates a proposal, "technical support procedures for stabilizing the network connection," and notifies the technical support representative (user). The user then provides specific technical support to the customer based on this proposal.

[0438] In this way, this system efficiently processes voice data from customers and enables quick and appropriate responses, thereby improving customer satisfaction and operational efficiency.

[0439] The processing flow will be explained below.

[0440] Step 1: Collecting audio data

[0441] The terminal receives calls and voice messages from customers. The received voice data is sent to the server in real time or in batch processing. The server stores this voice data in a database.

[0442] Step 2: Convert audio data to text

[0443] The server sends the voice data stored in the database to a voice recognition engine, which converts the voice data into text data, which is then stored in the database by the server.

[0444] Step 3: Analyze the text data

[0445] The server analyzes the text data using natural language processing (NLP) techniques. The server applies keyword extraction algorithms to extract important keywords. Furthermore, the server performs sentiment analysis to determine the sentiment in the text (e.g., positive, negative, neutral). Finally, the server performs topic identification to recognize the subject matter of the text data.

[0446] Step 4: Categorize

[0447] Based on the analysis, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[0448] Step 5: Allocate to products

[0449] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[0450] Step 6: Generate proposals

[0451] The server automatically generates suggestions based on the assigned text data. For example, if the data is related to technical support, it generates technical procedures for stabilizing network connections.

[0452] Step 7: Notification of proposal

[0453] The server sends the generated proposal to the notification system, which then notifies the relevant department (user) of the proposal. The user then checks the proposal and takes specific action.

[0454] The above processing steps enable accurate understanding of customer feedback and prompt and appropriate response.

[0455] Example 1

[0456] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0457] Conventional systems have difficulty efficiently collecting voice data from customers, appropriately converting it into text data, analyzing it, classifying it, and generating and notifying customers of proposals. In particular, the lack of keyword extraction, sentiment analysis, or topic identification during voice data analysis means that appropriate responses tend to be delayed. Furthermore, because proposals are not automatically generated using generative AI models, there is room for improvement in the accuracy and speed of responses. There is a need for a system that can resolve these issues and enable faster and more accurate responses.

[0458] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0459] In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data by extracting keywords, analyzing sentiment, and identifying topics using natural language processing technology, means for assigning the classified data to an appropriate department based on predefined categories, means for generating proposals using a generative AI model based on the assigned data, and means for notifying the generated proposals. This enables efficient and accurate processing of voice data from customers, enabling prompt and accurate responses.

[0460] "Customer voice data" refers to voice information transmitted by customers, including customer questions, complaints, requests, etc.

[0461] "Means for collection" refers to the device or software used to capture voice data from customers and input it into the system.

[0462] "Means for converting voice data into text data" refers to a device or software that converts voice data into text data using voice recognition technology.

[0463] "Means for analyzing text data" refers to a device or software that uses natural language processing techniques to analyze acquired text data.

[0464] "Means for classifying" refers to a device or software that appropriately classifies analyzed text data into predefined categories.

[0465] A "predefined category" is a category that is set in advance for classifying data, and examples include "technical support," "sales," and "complaints."

[0466] "Means for allocating to appropriate departments" refers to devices or software that transmit or distribute classified data to the departments corresponding to each category.

[0467] "Generative AI model" refers to an artificial intelligence model that uses deep learning technology to generate suggestions based on specific input data.

[0468] "Means for generating proposals" refers to devices or software that use generative AI models to automatically create proposals and solutions based on customer feedback.

[0469] "Means for notifying" refers to a device or software for communicating the generated proposal to an appropriate recipient.

[0470] A "prompt sentence" refers to text data that is input to a generative AI model to generate a specific output.

[0471] The present invention is a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies the customer of proposals. Specific embodiments are described below.

[0472] This system is mainly composed of three entities: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0473] Audio data collection

[0474] The terminal acquires voice data from the customer through a user-operated interface. Specifically, the terminal is equipped with a microphone and recording software (e.g., Audacity) that is used to collect voice data in real time or in batches. The collected voice data is encrypted and transmitted to a server using a secure communication protocol (e.g., HTTPS).

[0475] Converting audio data to text

[0476] The server stores the received voice data in a database. The server then sends the stored voice data to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice data into text data, which is then stored in the database again.

[0477] Text data analysis

[0478] The server analyzes the stored text data using natural language processing (NLP) techniques, specifically performing the following operations:

[0479] Keyword extraction: Uses the TF-IDF algorithm to extract key keywords from text data.

[0480] Sentiment analysis: Use a sentiment analysis API (e.g., IBM Watson Natural Language Understanding) to classify the sentiment of text data as positive, negative, or neutral.

[0481] Topic Identification: Identify topics within text data using pre-trained topic models (e.g., Latent Dirichlet Allocation).

[0482] Category sorting

[0483] Based on the results of the text data analysis, the server classifies the data into predefined categories (e.g., "technical support," "sales," "complaints") using the Naive Bayes classifier as the classification algorithm.

[0484] Allocation to products

[0485] The server assigns the classified text data to the appropriate department, using a department collaboration system (e.g., Slack or Microsoft Teams).

[0486] Proposal / proposal

[0487] The server automatically generates proposals using a generative AI model (e.g., OpenAI's GPT-3) based on the assigned data. The generated proposals are notified to the person in charge via a notification system (e.g., email or SMS gateway). The user receives the notification and takes specific action.

[0488] Specific examples

[0489] As a specific example, consider a case where voice data from a customer saying "The network connection is unstable" is collected by a terminal.

[0490] 1. The device collects the customer's voice message saying "The network connection is unstable," encrypts it, and sends it to the server.

[0491] 2. The server stores the voice data in a database and uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert it into text data such as "The network connection is unstable."

[0492] 3. The server analyzes this text data using natural language processing technology (e.g., IBM Watson Natural Language Understanding), extracts the keywords "network," "connection," and "unstable," and determines that the sentiment is negative.

[0493] 4. This data is classified under the category "Technical Support" and assigned to the Technical Support department.

[0494] 5. The server generates a suggestion by inputting the following prompt into the generative AI model (e.g., OpenAI's GPT-3): "The network connection is unstable. Please suggest technical support steps to resolve this."

[0495] 6. The generated proposal, "Technical support procedure for stabilizing network connections," is notified to the technical support staff. The technical support staff (user) provides specific technical support to the customer based on this proposal.

[0496] In this way, the system efficiently processes voice data from customers, enabling quick and accurate responses.

[0497] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0498] Step 1: Collecting audio data

[0499] Input: Customer voice input

[0500] Output: Audio data saved on the device (file format: .wav, etc.)

[0501] Specific operation:

[0502] The terminal utilizes a microphone and recording software (e.g., Audacity) to record audio in real time through the customer interface.

[0503] A customer says, "My network connection is unstable."

[0504] The device captures the audio and temporarily stores it in .wav format.

[0505] The saved audio data is encrypted and sent to the server.

[0506] Step 2: Convert audio data to text

[0507] Input: Collected encrypted audio data

[0508] Output: Text data

[0509] Specific operation:

[0510] The server stores the received voice data in a database.

[0511] The saved voice data is sent to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) and converted into text data.

[0512] The speech recognition engine returns text data saying "Network connection is unstable."

[0513] The server stores the converted text data in a database.

[0514] Step 3: Analyze the text data

[0515] Input: Text data

[0516] Output: Analysis results (keywords, sentiment, topics)

[0517] Specific operation:

[0518] The server analyzes the stored text data using natural language processing (NLP) technology.

[0519] Keyword extraction: The keywords “network”, “connection”, and “unstable” are extracted using the TF-IDF algorithm.

[0520] Sentiment analysis: Use a sentiment analysis API (e.g., IBM Watson Natural Language Understanding) to determine whether the text data is negative.

[0521] Topic Identification: Identify the topic "stabilizing network connectivity" using a pre-trained topic model (e.g., Latent Dirichlet Allocation).

[0522] Step 4: Categorize

[0523] Input: Analysis results (keywords, sentiment, topics)

[0524] Output: Classified category information

[0525] Specific operation:

[0526] Based on the analysis, the server classifies the data into predefined categories (e.g., "technical support").

[0527] The Naive Bayes classifier is used as the classification algorithm.

[0528] Information categorized under "Technical Support" is generated.

[0529] Step 5: Assign to departments

[0530] Input: Classified category information

[0531] Output: Data assigned to departments

[0532] Specific operation:

[0533] The server assigns data categorized under the "Technical Support" category to the Technical Support department.

[0534] Send data to the technical support department in real time using a cross-departmental system (e.g., Slack or Microsoft Teams).

[0535] Step 6: Proposal and project creation

[0536] Input: Data assigned to department

[0537] Output: Generated proposals

[0538] Specific operation:

[0539] The server generates suggestions using a generative AI model (e.g., OpenAI's GPT-3) based on the assigned data.

[0540] Enter the prompt sentence into the generative AI model: "The network connection is unstable. Please suggest technical support steps to resolve it."

[0541] The generative AI model generates "technical support procedures for stabilizing network connections," which are output as suggestions.

[0542] Step 7: Notification

[0543] Input: Generated proposal

[0544] Output: The suggestion sent to the user

[0545] Specific operation:

[0546] The server notifies the generated proposal to the technical support personnel via a notification system (e.g., email or SMS gateway).

[0547] The user receives a notification and takes specific action.

[0548] (Application example 1)

[0549] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0550] Current food delivery systems lack the means to quickly and accurately collect and analyze customer feedback. In particular, there is no system that can efficiently process voice feedback, which results in issues such as time spent improving customer satisfaction and resolving problems. Furthermore, conventional systems have difficulty accurately understanding customer sentiment and the content of feedback and quickly directing the appropriate department to respond.

[0551] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0552] In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data, means for assigning the classified data to a corresponding department, means for generating proposals based on the assigned data, means for notifying the generated proposals, means for extracting keywords from customer feedback using the voice data, and means for performing sentiment analysis using natural language processing technology to classify categories based on the keywords. This makes it possible to efficiently collect and analyze customer voice feedback and quickly and appropriately notify the corresponding department of proposals.

[0553] "Customer voice data" refers to voice information generated by customers, including feedback and inquiries about services and products.

[0554] "Means" refers to devices, methods, techniques or processes used to achieve a particular purpose.

[0555] "Text data" refers to information converted from audio into a string of characters, such as a collection of sentences or words.

[0556] "Keyword extraction" refers to the process of selecting key words and phrases from text data.

[0557] "Sentiment analysis" is the process of determining the writer's emotions from the content of text data, which are classified as positive, negative, neutral, etc.

[0558] "Natural language processing technology" refers to the technology that enables computers to understand, interpret, and generate human language, and includes keyword extraction and sentiment analysis as part of this.

[0559] A "department" refers to a specialized team responsible for a specific task or function within an organization.

[0560] "Proposals" refer to solutions or countermeasures that are automatically generated based on the analyzed data.

[0561] "Real-time" refers to data collection and processing occurring immediately, with minimal delay.

[0562] "Sentiment analysis using natural language processing technology" refers to the technology of using a computer to analyze emotions from text data, and is a part of natural language processing technology.

[0563] A "system" refers to an overall structure in which multiple components and means work together to perform a specific function.

[0564] This invention provides a system for food delivery services that efficiently collects and analyzes customer voice feedback and promptly notifies the relevant department of the proposed solution. This system consists of three entities: a server, a terminal, and a user.

[0565] System Overview

[0566] server

[0567] The server handles the central data processing: converting voice data to text, analyzing text data, allocating data to products, generating recommendations, and notifications.

[0568] Terminal

[0569] The device collects voice data from customers in real time. The data is picked up through a microphone and sent to a server. The device also serves as an interface for customers and guides them through the feedback collection process.

[0570] User

[0571] The user checks the proposals sent by the system and implements specific measures. The user refers to the person in charge of the relevant department, such as a technical support staff member or a manager.

[0572] Audio data collection

[0573] The device collects voice data from the microphone using the speech_recognition library. The voice information spoken by the customer is captured in real time and sent to the server. The collected voice data is converted to text data using the recognizer.recognize_google method.

[0574] Text data analysis

[0575] The server analyzes the acquired text data using natural language processing technology, specifically by performing the following operations:

[0576] Keyword Extraction: Use TextBlob to extract key words and phrases from text data.

[0577] Sentiment Analysis: Use TextBlob to classify the sentiment of text (positive, negative, neutral).

[0578] Category classification: Uses the transformers library pipeline to determine which category the text belongs to.

[0579] Product allocation and proposal generation

[0580] Based on the analysis results, the server assigns the data to the appropriate department and generates a recommendation for the corresponding department. For example, if the request is "feedback on an order," it will be assigned to the order processing department and a recommendation such as "Please check the order."

[0581] Notification of proposal details

[0582] The generated proposals are automatically notified to the relevant departments. The proposals are provided instantly through a notification system (pseudo implementation).

[0583] Specific examples

[0584] Below is an example of customer feedback:

[0585] "I couldn't specify a delivery time."

[0586] "The food tastes different than usual"

[0587] "The app is hard to understand"

[0588] For these specific examples, gather audio feedback using the following example prompts:

[0589] Example prompt sentence:

[0590] "Please provide your feedback by voice. For example, 'I can't confirm my order!' or 'The food tastes different.'"

[0591] Hardware and software used

[0592] Hardware: Microphone (for audio input)

[0593] Software: Python, speech_recognition, transformers, TextBlob

[0594] Data processing and calculation

[0595] The server converts the voice data into text and performs keyword extraction, sentiment analysis, and categorization, allowing it to properly categorize customer feedback and quickly notify the relevant department with suggestions.

[0596] In this way, the present invention provides a system that improves customer satisfaction in food delivery services and also improves operational efficiency.

[0597] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0598] Step 1:

[0599] The terminal collects voice data. Specifically, it obtains voice feedback from the customer through a microphone and collects it in real time using a speech recognition engine (speech_recognition). The input is the customer's voice, and the output is voice data.

[0600] Step 2:

[0601] The terminal transmits the collected voice data to the server. The input is the voice data, and the output is the voice data transmitted to the server.

[0602] Step 3:

[0603] The server converts the received voice data into text data. Specifically, it converts the voice data into text using the recognizer.recognize_google method. The input is voice data, and the output is text data.

[0604] Step 4:

[0605] The server analyzes the converted text data. Specifically, it extracts keywords from the text data (using TextBlob), performs sentiment analysis (also using TextBlob), and classifies it into appropriate categories (using a pipeline from the transformers library). The input is text data, and the output is keywords, sentiment classification, and categorized data.

[0606] Step 5:

[0607] The server assigns data to the corresponding department based on the analysis results. The input is keywords, sentiment classification, and categorized data, and the output is the assignment of data to the corresponding department.

[0608] Step 6:

[0609] The server generates proposals based on the assigned data. Specifically, it automatically proposes predefined countermeasures based on the analyzed categories. The input is the data assigned to the corresponding department, and the output is the proposals.

[0610] Step 7:

[0611] The server notifies the person in charge of the corresponding department of the content of the generated proposal via the notification system. The input is the content of the proposal, and the output is the notification to the person in charge.

[0612] Step 8:

[0613] The user (person in charge) implements specific measures based on the proposal. The input is the proposal, and the output is the implementation of specific measures.

[0614] This series of steps enables efficient processing of voice feedback from customers and provides prompt and appropriate responses.

[0615] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0616] The present invention relates to a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies proposals. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to respond more accurately. Specific embodiments of this system will be described below.

[0617] System Overview

[0618] This system consists of three components: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0619] Audio data collection

[0620] The terminal captures voice data from the customer in real time or in batch processing, and the captured voice data is sent to the server and stored in a database.

[0621] Converting audio data to text

[0622] The server sends the saved voice data to a voice recognition engine, which converts the voice data into text data, which is then saved in a database by the server.

[0623] Text data analysis

[0624] The server analyzes the text data using natural language processing (NLP) techniques and an emotion engine. Specifically, this process involves the following operations:

[0625] Keyword extraction: Using specific algorithms to extract key keywords from text data.

[0626] Sentiment analysis: Using a sentiment engine to determine user sentiment from the content of text data (e.g., positive, negative, neutral).

[0627] Topic identification: Identifying themes and topics based on text data.

[0628] Category sorting

[0629] Based on the analysis results, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[0630] Allocation to products

[0631] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[0632] Proposal / proposal

[0633] The server automatically generates suggestions based on the assigned text data. The generated suggestions reflect the emotions determined by the emotion engine. For example, if negative emotions are included, the server will suggest response procedures that require special attention.

[0634] Notification of proposal details

[0635] The server sends the generated proposal to the notification system, which then notifies the relevant department (user) of the proposal. The user then checks the proposal and takes specific action.

[0636] Specific examples

[0637] As a concrete example, consider the case of processing voice data from a customer saying, "The network connection is unstable." The device collects this voice data, and the server uses a speech recognition engine to convert it into text data saying, "The network connection is unstable." The server then analyzes this text data using NLP technology and an emotion engine to extract the keywords "network," "connection," and "unstable," and determines that the emotion is negative. This data is classified into the "technical support" category and assigned to the technical support department. Finally, the server generates a proposal saying, "Technical support procedures for stabilizing the network connection, and particularly, a prompt response is required," and notifies the technical support staff (user). The user then provides specific technical support to the customer based on this proposal.

[0638] In this way, this system efficiently processes voice data from customers and, by combining it with an emotion engine, provides more accurate responses, enabling quick and appropriate responses.

[0639] The processing flow will be explained below.

[0640] Step 1: Collecting audio data

[0641] The terminal receives calls and voice messages from customers in real time, and the received voice data is sent from the terminal to a server, which stores the voice data in a database.

[0642] Step 2: Convert audio data to text

[0643] The server sends the voice data stored in the database to a voice recognition engine, which converts the voice data into text data, which is then stored in the database by the server.

[0644] Step 3: Analyze the text data

[0645] The server analyzes the text data using natural language processing (NLP) technology and an emotion engine. Specifically, it uses NLP technology to perform keyword extraction and identify important keywords. It then uses the emotion engine to determine the sentiment (positive, negative, neutral) in the text data. It also performs topic identification and identifies the main themes in the text data.

[0646] Step 4: Categorize

[0647] Based on the analysis, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[0648] Step 5: Allocate to products

[0649] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[0650] Step 6: Generate proposals

[0651] The server automatically generates suggestions based on the assigned text data. The suggestions reflect the emotions determined by the emotion engine. For example, if negative emotions are included, the server will suggest procedures that require special attention.

[0652] Step 7: Notification of proposal

[0653] The server sends the generated proposal to the notification system, which notifies the person in charge (user) of the proposal.

[0654] Step 8: User Action

[0655] The user receives the proposal from the notification system and takes specific action based on the proposal. For example, in the case of technical support, the user provides technical support to the customer based on the proposal.

[0656] The above processing steps enable the system to accurately understand customer feedback and respond quickly and appropriately. By combining this system with an emotion engine, the quality of responses can be further improved.

[0657] Example 2

[0658] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0659] Conventional customer support systems had problems in efficiently collecting and analyzing voice data from customers, allocating it to the appropriate department based on its content, and quickly generating and notifying proposals. In particular, they lacked the ability to respond in a way that took customer emotions into consideration, making it difficult to improve customer satisfaction.

[0660] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0661] In this invention, the server includes a means for the terminal to collect voice data in real time or batch processing and send it to the server, a means for the server to convert the voice data into text data using a voice recognition engine, and a means for the server to analyze the text data using natural language processing technology and an emotion engine. This makes it possible to efficiently collect and analyze voice data from customers, assign it to the appropriate department based on its content, and quickly generate and notify proposals. It also enables responses that take customer emotions into consideration, which is expected to improve customer satisfaction.

[0662] "Customer Voice Data" refers to information in voice format transmitted by a Customer.

[0663] "Collection means" refers to a combination of hardware and software for capturing and recording audio data.

[0664] "Means for converting into text data" refers to technology and devices that analyze voice data and convert it into corresponding character string information.

[0665] "Means of analysis and classification" refers to the process of processing text data, analyzing its content, and classifying it into specific categories.

[0666] "Means for allocating to the corresponding department" refers to a system or algorithm for transferring classified data to the most appropriate department.

[0667] "Means for generating proposal content" refers to technology that automatically creates countermeasures and proposal content based on analyzed data.

[0668] "Means for notifying" refers to communication devices and systems for notifying the relevant departments and personnel of the content of the generated proposal.

[0669] "Real-time or batch processing" refers to a method of processing data immediately and a method of processing data in bulk.

[0670] "Speech recognition engine" refers to a software and hardware system for converting voice data into text data.

[0671] "Natural language processing technology" refers to computer technologies and algorithms for understanding, analyzing, and generating text data.

[0672] An "emotion engine" refers to software and algorithms that analyze text data and determine emotions from it.

[0673] MODE FOR CARRYING OUT THE INVENTION

[0674] This invention relates to a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies proposals. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to respond more accurately. Specific embodiments are described below.

[0675] This system consists of three components: a server, a terminal, and a user. The server processes the main data, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0676] Audio data collection

[0677] The terminal captures voice data from the customer in real time or in batch processing using a microphone device and recording software (e.g., Audacity). The captured voice data is stored in temporary storage and later transmitted to the server.

[0678] Converting audio data to text

[0679] The server converts the received voice data into text data using a speech recognition engine (for example, Google Cloud Speech-to-Text API), and the converted text data is stored in a database.

[0680] Text data analysis

[0681] The server analyzes the text data using natural language processing technology and an emotion engine (e.g., Hume AI's emotion recognition API). The server extracts keywords using the TF-IDF algorithm and determines the user's emotion (positive, negative, neutral, etc.) using the emotion engine. It also uses natural language processing technology to identify topics.

[0682] Category sorting

[0683] The server then classifies the text data into predefined categories based on the analysis results (e.g., "technical support," "sales," "complaint," etc.) using the text classification capabilities of the Google Cloud Natural Language API.

[0684] Allocation to departments

[0685] The server assigns text data that has been classified into the appropriate category to the corresponding department, for example, data classified as "technical support" will be sent to the technical support department.

[0686] Proposal / proposal

[0687] The server automatically generates suggestions based on the assigned text data. The suggestions reflect the emotions determined by the emotion engine, and if negative emotions are included, a quick response is required.

[0688] Notification of proposal details

[0689] The server sends the generated proposal to the notification system. The notification system notifies the person in charge (user) of the proposal. The user confirms the notified proposal and takes specific action.

[0690] Specific examples

[0691] For example, consider the case where voice data from a customer saying "The network connection is unstable" is processed.

[0692] 1. The device collects and records this audio data.

[0693] 2. The server sends the audio file to the Google Cloud Speech-to-Text API and receives the text "Network connection is unstable."

[0694] 3. The server analyzes this text data using NLP technology and an emotion engine, extracts the keywords "network," "connection," and "unstable," and determines that the emotion is negative.

[0695] 4. The server categorizes this text data into the "Technical Support" category.

[0696] 5. The server sends this data to the technical support department.

[0697] 6. The server generates a suggestion: "Technical support procedures for stabilizing network connectivity and particularly rapid response is required."

[0698] 7. The server sends the generated proposal to the notification system, which then notifies the technical support staff (user). The user then provides specific technical support to the customer based on the proposal.

[0699] In this way, the system efficiently processes voice data from customers and, by combining it with an emotion engine, provides more accurate responses, enabling quick and appropriate responses.

[0700] Prompt Sentence Examples

[0701] For example, a prompt to a generative AI model might look like this:

[0702] "Convert the customer's voice data, 'The network connection is unstable,' into text data using NLP technology and a sentiment analysis engine, extract keywords, determine sentiment, and convert it into specific suggestions."

[0703] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0704] Step 1:

[0705] The terminal collects voice data in real time or in batch processing. It uses a microphone device and recording software (e.g., Audacity) to capture voice data from customers. This saves the customer's voice as an audio file, which is then sent to the server.

[0706] Specific operation: When a customer says "The network connection is unstable," the device's microphone records this, saves it as an audio file, and sends it to the server.

[0707] Input: Customer voice data

[0708] Output: Audio file

[0709] Step 2:

[0710] The server sends the received audio file to a speech recognition engine, which uses the Google Cloud Speech-to-Text API to convert the audio data into text data, which the server then stores in a database.

[0711] Specific operation: The server sends the audio file "The network connection is unstable" to the Google Cloud Speech-to-Text API, obtains the text data "The network connection is unstable", and stores it in the database.

[0712] Input: Audio file

[0713] Output: Text data

[0714] Step 3:

[0715] The server analyzes the stored text data using natural language processing technology and an emotion engine (for example, Hume AI's emotion recognition API). Specifically, it uses the TF-IDF algorithm to extract keywords, and the emotion engine determines the user's emotion (positive, negative, neutral, etc.). It also uses natural language processing technology to identify topics.

[0716] Specific operation: The server analyzes the text data "Your network connection is unstable," extracts the keywords "network," "connection," and "unstable," and determines that the content contains negative sentiment.

[0717] Input: Text data

[0718] Output: Keywords, sentiment analysis, identified topics

[0719] Step 4:

[0720] The server classifies the text data into predefined categories based on the analysis results, such as "technical support," "sales," and "complaints," using the Google Cloud Natural Language API.

[0721] Specific operation: The server classifies the text data "Network connection is unstable" into the "Technical Support" category.

[0722] Input: Keywords, sentiment analysis results, identified topics

[0723] Output: Categorized data

[0724] Step 5:

[0725] The server assigns text data that has been classified into the appropriate category to the corresponding department, for example, data classified as "technical support" is sent to the technical support department.

[0726] What happens: The server sends data categorized under "Tech Support" to the technical support department.

[0727] Input: Categorical data

[0728] Output: Data assigned to departments

[0729] Step 6:

[0730] The server automatically generates suggestions based on the assigned text data, taking into account the emotions (e.g., negative) determined by the emotion engine.

[0731] What happens: The server generates a suggestion that states, "Technical support procedures for stabilizing network connectivity, and particularly prompt action, are needed."

[0732] Input: Data assigned to department, emotion judgment result

[0733] Output: Proposal

[0734] Step 7:

[0735] The server sends the generated proposal to the notification system. The notification system notifies the person in charge (user) of the proposal. The user confirms the notified proposal and takes specific action.

[0736] Specific operation: The server sends the proposal to the notification system, which notifies the technical support staff (user). The user provides specific technical support based on the proposal.

[0737] Input: Proposal

[0738] Output: Proposal notified

[0739] (Application example 2)

[0740] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0741] Conventional customer service systems have had the problem of being unable to efficiently collect customer feedback, quickly analyze and classify it, and make appropriate proposals. In particular, when dealing with customers in physical stores, real-time feedback is required, but conventional systems were inadequate in this regard.

[0742] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data, means for assigning the classified data to the corresponding department, means for generating proposals based on the assigned data, means for notifying the generated proposals, and smartphone application means for responding to customers in real time at the store. This makes it possible to collect and analyze customer feedback in real time at a physical store and quickly make appropriate proposals.

[0743] "Customer voice data" refers to voice information of words and phrases spoken by a customer.

[0744] "Means of collection" refers to devices or systems for capturing audio data, such as microphones or voice recording applications.

[0745] "Means for converting into text data" refers to a voice recognition engine or software for converting voice data into text information.

[0746] "Means for analyzing and classifying" refers to natural language processing techniques and algorithms used to analyze the converted text data and classify it into specific categories.

[0747] The "means for allocating to the corresponding department" refers to a system, such as a database management system, that automatically sorts the analyzed and classified data to the appropriate business department.

[0748] "Means for generating recommendations" refers to software algorithms that automatically generate solutions or advice based on the assigned data.

[0749] "Means of notification" refers to messaging and alert systems that notify relevant department staff of generated suggestions and countermeasures.

[0750] "Customer service in stores" refers to the business activities of responding quickly to customer requests and questions in physical stores.

[0751] "Smartphone application means" refers to dedicated application software that allows users to respond to customers via their smartphones.

[0752] The system for implementing this invention is composed of three entities: a server, a terminal, and a user. The specific roles and processing contents of each entity are explained below.

[0753] Audio data collection and text conversion

[0754] The terminal uses a dedicated application installed on a smartphone to handle customer inquiries in real time within the store. The smartphone's microphone is used to collect customer voice data. This voice data is then sent to a server, where it is converted into text data using a speech recognition engine (for example, Google's speech recognition API). The converted text data is then stored on the server.

[0755] Analysis and classification of text data

[0756] The server then analyzes the converted text data using natural language processing (NLP) techniques, such as Hugging Face's BERT model, to extract keywords, analyze sentiment, and identify topics from the text data. It also uses an emotion engine to determine the user's sentiment (e.g., positive, negative, neutral) based on the content of the text data. Based on the analysis results, the text data is classified into predefined categories (e.g., technical support, sales, complaints).

[0757] Allocating data and generating proposals

[0758] The server automatically assigns the classified text data to the appropriate department. For example, data classified as "technical support" is sent to the technical support department. Next, a suggestion is automatically generated. The generated suggestion reflects the emotion determined by the emotion engine. If a negative emotion is included, a response procedure requiring special attention is suggested.

[0759] Notification and implementation of proposals

[0760] The server sends the generated proposal to a notification system, which notifies staff in the relevant departments. Notification methods include push notifications via smartphone applications and internal messaging systems. The users (staff) then check the notified proposal and implement specific measures. This enables quick and appropriate responses that reflect customer feedback in real time.

[0761] Adding specific examples

[0762] As a concrete example, consider the case where a customer in a store says, "I don't know where the product is." This voice data is collected by a smartphone app and converted into text data, "I don't know where the product is," using a voice recognition engine. The server then analyzes this text data using NLP technology and an emotion engine, extracting the keywords "product," "location," and "don't know," and determining that the emotion is negative. This data is classified into the "store support" category and assigned to the relevant department. Finally, the server generates a suggestion, "You need guidance to the product location and prompt customer service," and notifies the staff.

[0763] Prompt Sentence Examples

[0764] Voice data is collected and converted into text data. The converted text data is analyzed using NLP technology to extract keywords, analyze sentiment, and identify topics. Based on the results, appropriate suggestions are generated and notified to the relevant departments.

[0765] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0766] Step 1: Collecting audio data

[0767] The device (smartphone) collects the customer's voice data through the application using a microphone. The input is the customer's voice data, and the output is the collected voice file. This voice file is sent directly to the server.

[0768] Step 2: Convert audio data to text

[0769] The server receives the transmitted audio file and converts it into text data using a speech recognition engine. The input is the audio file, and the output is the converted text data. Specifically, the server analyzes the audio data using Google's speech recognition API and generates the corresponding text data.

[0770] Step 3: Save the text data

[0771] The server stores the converted text data in a database. The input of this process is the text data obtained in the previous step, and the output is the stored text data. A database management system is used to efficiently store the text data.

[0772] Step 4: Analyze and classify text data

[0773] The server analyzes the stored text data using natural language processing (NLP) technology and a sentiment analysis engine. Specifically, it uses the Hugging Face BERT model to extract keywords, analyze sentiment, and identify topics from the text data. The input is the text data, and the output is the extracted keywords, sentiment classification, and topic identification results.

[0774] Step 5: Allocate data

[0775] The server automatically assigns data to the appropriate department based on the analyzed and classified data. The input is keywords, sentiment classification, and topic identification results, and the output is department assignment information. This is done based on pre-defined categories.

[0776] Step 6: Generate proposals

[0777] The server generates appropriate proposals based on the assigned data. The input is the assignment information to the corresponding department, and the output is the generated proposal. The proposals are automatically created using an algorithm and include appropriate countermeasures.

[0778] Step 7: Notification of proposal

[0779] The server sends the generated proposal to a notification system, which notifies staff in the relevant departments. The input is the proposal, and the output is the notified information. Notification methods include push notifications on smartphones and internal messaging systems.

[0780] Step 8: Implementing customer interactions

[0781] The user (staff member) checks the notified proposal and takes specific action. The input is the notified proposal, and the output is the result of the action taken. This enables quick and appropriate action that reflects customer feedback in real time.

[0782] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0783] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0784] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0785] [Third embodiment]

[0786] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0787] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0788] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0789] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0790] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0791] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0792] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0793] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0794] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0795] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0796] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0797] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0798] This invention is a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies the customer of proposals. A specific embodiment of this system will be described.

[0799] System Overview

[0800] This system is mainly composed of three components: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0801] Audio data collection

[0802] The terminal captures voice data from the customer in real time or in batch processing, and the captured voice data is sent to the server and stored in a database.

[0803] Converting audio data to text

[0804] The server sends the saved voice data to a voice recognition engine, which converts the voice data into text data and saves the text data back in the database.

[0805] Text data analysis

[0806] The server analyzes the text data using natural language processing (NLP) techniques, which specifically perform the following operations:

[0807] Keyword extraction: Using specific algorithms to extract key keywords from text data.

[0808] Sentiment analysis: Determine the sentiment (positive, negative, neutral) from the content of text data.

[0809] Topic identification: Identifying themes and topics based on text data.

[0810] Category sorting

[0811] The server analyzes the text data and classifies it into predefined categories, such as "technical support," "sales," and "complaints."

[0812] Allocation to products

[0813] The server assigns categorized text data to the appropriate department, for example, data categorized in the "Tech Support" category is sent to the Tech Support department.

[0814] Proposal / proposal

[0815] The server automatically generates proposals based on the assigned data. The generated proposals are notified to the person in charge (user) of the corresponding department via the notification system. The user then checks the notified proposals and takes specific action.

[0816] Specific examples

[0817] As a concrete example, consider the case of processing voice data from a customer saying, "The network connection is unstable." The device collects this voice data, and the server uses a speech recognition engine to convert it into text data saying, "The network connection is unstable." The server then analyzes this text data using NLP technology to extract the keywords "network," "connection," and "unstable," and determines that the sentiment is negative. This data is classified into the "technical support" category and assigned to the technical support department. Finally, the server generates a proposal, "technical support procedures for stabilizing the network connection," and notifies the technical support representative (user). The user then provides specific technical support to the customer based on this proposal.

[0818] In this way, this system efficiently processes voice data from customers and enables quick and appropriate responses, thereby improving customer satisfaction and operational efficiency.

[0819] The processing flow will be explained below.

[0820] Step 1: Collecting audio data

[0821] The terminal receives calls and voice messages from customers. The received voice data is sent to the server in real time or in batch processing. The server stores this voice data in a database.

[0822] Step 2: Convert audio data to text

[0823] The server sends the voice data stored in the database to a voice recognition engine, which converts the voice data into text data, which is then stored in the database by the server.

[0824] Step 3: Analyze the text data

[0825] The server analyzes the text data using natural language processing (NLP) techniques. The server applies keyword extraction algorithms to extract important keywords. Furthermore, the server performs sentiment analysis to determine the sentiment in the text (e.g., positive, negative, neutral). Finally, the server performs topic identification to recognize the subject matter of the text data.

[0826] Step 4: Categorize

[0827] Based on the analysis, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[0828] Step 5: Allocate to products

[0829] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[0830] Step 6: Generate proposals

[0831] The server automatically generates suggestions based on the assigned text data. For example, if the data is related to technical support, it generates technical procedures for stabilizing network connections.

[0832] Step 7: Notification of proposal

[0833] The server sends the generated proposal to the notification system, which then notifies the relevant department (user) of the proposal. The user then checks the proposal and takes specific action.

[0834] The above processing steps enable accurate understanding of customer feedback and prompt and appropriate response.

[0835] Example 1

[0836] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0837] Conventional systems have difficulty efficiently collecting voice data from customers, appropriately converting it into text data, analyzing it, classifying it, and generating and notifying customers of proposals. In particular, the lack of keyword extraction, sentiment analysis, or topic identification during voice data analysis means that appropriate responses tend to be delayed. Furthermore, because proposals are not automatically generated using generative AI models, there is room for improvement in the accuracy and speed of responses. There is a need for a system that can resolve these issues and enable faster and more accurate responses.

[0838] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0839] In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data by extracting keywords, analyzing sentiment, and identifying topics using natural language processing technology, means for assigning the classified data to an appropriate department based on predefined categories, means for generating proposals using a generative AI model based on the assigned data, and means for notifying the generated proposals. This enables efficient and accurate processing of voice data from customers, enabling prompt and accurate responses.

[0840] "Customer voice data" refers to voice information transmitted by customers, including customer questions, complaints, requests, etc.

[0841] "Means for collection" refers to the device or software used to capture voice data from customers and input it into the system.

[0842] "Means for converting voice data into text data" refers to a device or software that converts voice data into text data using voice recognition technology.

[0843] "Means for analyzing text data" refers to a device or software that uses natural language processing techniques to analyze acquired text data.

[0844] "Means for classifying" refers to a device or software that appropriately classifies analyzed text data into predefined categories.

[0845] A "predefined category" is a category that is set in advance for classifying data, and examples include "technical support," "sales," and "complaints."

[0846] "Means for allocating to appropriate departments" refers to devices or software that transmit or distribute classified data to the departments corresponding to each category.

[0847] "Generative AI model" refers to an artificial intelligence model that uses deep learning technology to generate suggestions based on specific input data.

[0848] "Means for generating proposals" refers to devices or software that use generative AI models to automatically create proposals and solutions based on customer feedback.

[0849] "Means for notifying" refers to a device or software for communicating the generated proposal to an appropriate recipient.

[0850] A "prompt sentence" refers to text data that is input to a generative AI model to generate a specific output.

[0851] The present invention is a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies the customer of proposals. Specific embodiments are described below.

[0852] This system is mainly composed of three entities: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0853] Audio data collection

[0854] The terminal acquires voice data from the customer through a user-operated interface. Specifically, the terminal is equipped with a microphone and recording software (e.g., Audacity) that is used to collect voice data in real time or in batches. The collected voice data is encrypted and transmitted to a server using a secure communication protocol (e.g., HTTPS).

[0855] Converting audio data to text

[0856] The server stores the received voice data in a database. The server then sends the stored voice data to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice data into text data, which is then stored in the database again.

[0857] Text data analysis

[0858] The server analyzes the stored text data using natural language processing (NLP) techniques, specifically performing the following operations:

[0859] Keyword extraction: Uses the TF-IDF algorithm to extract key keywords from text data.

[0860] Sentiment analysis: Use a sentiment analysis API (e.g., IBM Watson Natural Language Understanding) to classify the sentiment of text data as positive, negative, or neutral.

[0861] Topic Identification: Identify topics within text data using pre-trained topic models (e.g., Latent Dirichlet Allocation).

[0862] Category sorting

[0863] Based on the results of the text data analysis, the server classifies the data into predefined categories (e.g., "technical support," "sales," "complaints") using the Naive Bayes classifier as the classification algorithm.

[0864] Allocation to products

[0865] The server assigns the classified text data to the appropriate department, using a department collaboration system (e.g., Slack or Microsoft Teams).

[0866] Proposal / proposal

[0867] The server automatically generates proposals using a generative AI model (e.g., OpenAI's GPT-3) based on the assigned data. The generated proposals are notified to the person in charge via a notification system (e.g., email or SMS gateway). The user receives the notification and takes specific action.

[0868] Specific examples

[0869] As a specific example, consider a case where voice data from a customer saying "The network connection is unstable" is collected by a terminal.

[0870] 1. The device collects the customer's voice message saying "The network connection is unstable," encrypts it, and sends it to the server.

[0871] 2. The server stores the voice data in a database and uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert it into text data such as "The network connection is unstable."

[0872] 3. The server analyzes this text data using natural language processing technology (e.g., IBM Watson Natural Language Understanding), extracts the keywords "network," "connection," and "unstable," and determines that the sentiment is negative.

[0873] 4. This data is classified under the category "Technical Support" and assigned to the Technical Support department.

[0874] 5. The server generates a suggestion by inputting the following prompt into the generative AI model (e.g., OpenAI's GPT-3): "The network connection is unstable. Please suggest technical support steps to resolve this."

[0875] 6. The generated proposal, "Technical support procedure for stabilizing network connections," is notified to the technical support staff. The technical support staff (user) provides specific technical support to the customer based on this proposal.

[0876] In this way, the system efficiently processes voice data from customers, enabling quick and accurate responses.

[0877] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0878] Step 1: Collecting audio data

[0879] Input: Customer voice input

[0880] Output: Audio data saved on the device (file format: .wav, etc.)

[0881] Specific operation:

[0882] The terminal utilizes a microphone and recording software (e.g., Audacity) to record audio in real time through the customer interface.

[0883] A customer says, "My network connection is unstable."

[0884] The device captures the audio and temporarily stores it in .wav format.

[0885] The saved audio data is encrypted and sent to the server.

[0886] Step 2: Convert audio data to text

[0887] Input: Collected encrypted audio data

[0888] Output: Text data

[0889] Specific operation:

[0890] The server stores the received voice data in a database.

[0891] The saved voice data is sent to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) and converted into text data.

[0892] The speech recognition engine returns text data saying "Network connection is unstable."

[0893] The server stores the converted text data in a database.

[0894] Step 3: Analyze the text data

[0895] Input: Text data

[0896] Output: Analysis results (keywords, sentiment, topics)

[0897] Specific operation:

[0898] The server analyzes the stored text data using natural language processing (NLP) technology.

[0899] Keyword extraction: The keywords “network”, “connection”, and “unstable” are extracted using the TF-IDF algorithm.

[0900] Sentiment analysis: Use a sentiment analysis API (e.g., IBM Watson Natural Language Understanding) to determine whether the text data is negative.

[0901] Topic Identification: Identify the topic "stabilizing network connectivity" using a pre-trained topic model (e.g., Latent Dirichlet Allocation).

[0902] Step 4: Categorize

[0903] Input: Analysis results (keywords, sentiment, topics)

[0904] Output: Classified category information

[0905] Specific operation:

[0906] Based on the analysis, the server classifies the data into predefined categories (e.g., "technical support").

[0907] The Naive Bayes classifier is used as the classification algorithm.

[0908] Information categorized under "Technical Support" is generated.

[0909] Step 5: Assign to departments

[0910] Input: Classified category information

[0911] Output: Data assigned to departments

[0912] Specific operation:

[0913] The server assigns data categorized under the "Technical Support" category to the Technical Support department.

[0914] Send data to the technical support department in real time using a cross-departmental system (e.g., Slack or Microsoft Teams).

[0915] Step 6: Proposal and project creation

[0916] Input: Data assigned to department

[0917] Output: Generated proposals

[0918] Specific operation:

[0919] The server generates suggestions using a generative AI model (e.g., OpenAI's GPT-3) based on the assigned data.

[0920] Enter the prompt sentence into the generative AI model: "The network connection is unstable. Please suggest technical support steps to resolve it."

[0921] The generative AI model generates "technical support procedures for stabilizing network connections," which are output as suggestions.

[0922] Step 7: Notification

[0923] Input: Generated proposal

[0924] Output: The suggestion sent to the user

[0925] Specific operation:

[0926] The server notifies the generated proposal to the technical support personnel via a notification system (e.g., email or SMS gateway).

[0927] The user receives a notification and takes specific action.

[0928] (Application example 1)

[0929] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0930] Current food delivery systems lack the means to quickly and accurately collect and analyze customer feedback. In particular, there is no system that can efficiently process voice feedback, which results in issues such as time spent improving customer satisfaction and resolving problems. Furthermore, conventional systems have difficulty accurately understanding customer sentiment and the content of feedback and quickly directing the appropriate department to respond.

[0931] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0932] In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data, means for assigning the classified data to a corresponding department, means for generating proposals based on the assigned data, means for notifying the generated proposals, means for extracting keywords from customer feedback using the voice data, and means for performing sentiment analysis using natural language processing technology to classify categories based on the keywords. This makes it possible to efficiently collect and analyze customer voice feedback and quickly and appropriately notify the corresponding department of proposals.

[0933] "Customer voice data" refers to voice information generated by customers, including feedback and inquiries about services and products.

[0934] "Means" refers to devices, methods, techniques or processes used to achieve a particular purpose.

[0935] "Text data" refers to information converted from audio into a string of characters, such as a collection of sentences or words.

[0936] "Keyword extraction" refers to the process of selecting key words and phrases from text data.

[0937] "Sentiment analysis" is the process of determining the writer's emotions from the content of text data, which are classified as positive, negative, neutral, etc.

[0938] "Natural language processing technology" refers to the technology that enables computers to understand, interpret, and generate human language, and includes keyword extraction and sentiment analysis as part of this.

[0939] A "department" refers to a specialized team responsible for a specific task or function within an organization.

[0940] "Proposals" refer to solutions or countermeasures that are automatically generated based on the analyzed data.

[0941] "Real-time" refers to data collection and processing occurring immediately, with minimal delay.

[0942] "Sentiment analysis using natural language processing technology" refers to the technology of using a computer to analyze emotions from text data, and is a part of natural language processing technology.

[0943] A "system" refers to an overall structure in which multiple components and means work together to perform a specific function.

[0944] This invention provides a system for food delivery services that efficiently collects and analyzes customer voice feedback and promptly notifies the relevant department of the proposed solution. This system consists of three entities: a server, a terminal, and a user.

[0945] System Overview

[0946] server

[0947] The server handles the central data processing: converting voice data to text, analyzing text data, allocating data to products, generating recommendations, and notifications.

[0948] Terminal

[0949] The device collects voice data from customers in real time. The data is picked up through a microphone and sent to a server. The device also serves as an interface for customers and guides them through the feedback collection process.

[0950] User

[0951] The user checks the proposals sent by the system and implements specific measures. The user refers to the person in charge of the relevant department, such as a technical support staff member or a manager.

[0952] Audio data collection

[0953] The device collects voice data from the microphone using the speech_recognition library. The voice information spoken by the customer is captured in real time and sent to the server. The collected voice data is converted to text data using the recognizer.recognize_google method.

[0954] Text data analysis

[0955] The server analyzes the acquired text data using natural language processing technology, specifically by performing the following operations:

[0956] Keyword Extraction: Use TextBlob to extract key words and phrases from text data.

[0957] Sentiment Analysis: Use TextBlob to classify the sentiment of text (positive, negative, neutral).

[0958] Category classification: Uses the transformers library pipeline to determine which category the text belongs to.

[0959] Product allocation and proposal generation

[0960] Based on the analysis results, the server assigns the data to the appropriate department and generates a recommendation for the corresponding department. For example, if the request is "feedback on an order," it will be assigned to the order processing department and a recommendation such as "Please check the order."

[0961] Notification of proposal details

[0962] The generated proposals are automatically notified to the relevant departments. The proposals are provided instantly through a notification system (pseudo implementation).

[0963] Specific examples

[0964] Below is an example of customer feedback:

[0965] "I couldn't specify a delivery time."

[0966] "The food tastes different than usual"

[0967] "The app is hard to understand"

[0968] For these specific examples, gather audio feedback using the following example prompts:

[0969] Example prompt sentence:

[0970] "Please provide your feedback by voice. For example, 'I can't confirm my order!' or 'The food tastes different.'"

[0971] Hardware and software used

[0972] Hardware: Microphone (for audio input)

[0973] Software: Python, speech_recognition, transformers, TextBlob

[0974] Data processing and calculation

[0975] The server converts the voice data into text and performs keyword extraction, sentiment analysis, and categorization, allowing it to properly categorize customer feedback and quickly notify the relevant department with suggestions.

[0976] In this way, the present invention provides a system that improves customer satisfaction in food delivery services and also improves operational efficiency.

[0977] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0978] Step 1:

[0979] The terminal collects voice data. Specifically, it obtains voice feedback from the customer through a microphone and collects it in real time using a speech recognition engine (speech_recognition). The input is the customer's voice, and the output is voice data.

[0980] Step 2:

[0981] The terminal transmits the collected voice data to the server. The input is the voice data, and the output is the voice data transmitted to the server.

[0982] Step 3:

[0983] The server converts the received voice data into text data. Specifically, it converts the voice data into text using the recognizer.recognize_google method. The input is voice data, and the output is text data.

[0984] Step 4:

[0985] The server analyzes the converted text data. Specifically, it extracts keywords from the text data (using TextBlob), performs sentiment analysis (also using TextBlob), and classifies it into appropriate categories (using a pipeline from the transformers library). The input is text data, and the output is keywords, sentiment classification, and categorized data.

[0986] Step 5:

[0987] The server assigns data to the corresponding department based on the analysis results. The input is keywords, sentiment classification, and categorized data, and the output is the assignment of data to the corresponding department.

[0988] Step 6:

[0989] The server generates proposals based on the assigned data. Specifically, it automatically proposes predefined countermeasures based on the analyzed categories. The input is the data assigned to the corresponding department, and the output is the proposals.

[0990] Step 7:

[0991] The server notifies the person in charge of the corresponding department of the content of the generated proposal via the notification system. The input is the content of the proposal, and the output is the notification to the person in charge.

[0992] Step 8:

[0993] The user (person in charge) implements specific measures based on the proposal. The input is the proposal, and the output is the implementation of specific measures.

[0994] This series of steps enables efficient processing of voice feedback from customers and provides prompt and appropriate responses.

[0995] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0996] The present invention relates to a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies proposals. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to respond more accurately. Specific embodiments of this system will be described below.

[0997] System Overview

[0998] This system consists of three components: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[0999] Audio data collection

[1000] The terminal captures voice data from the customer in real time or in batch processing, and the captured voice data is sent to the server and stored in a database.

[1001] Converting audio data to text

[1002] The server sends the saved voice data to a voice recognition engine, which converts the voice data into text data, which is then saved in a database by the server.

[1003] Text data analysis

[1004] The server analyzes the text data using natural language processing (NLP) techniques and an emotion engine. Specifically, this process involves the following operations:

[1005] Keyword extraction: Using specific algorithms to extract key keywords from text data.

[1006] Sentiment analysis: Using a sentiment engine to determine user sentiment from the content of text data (e.g., positive, negative, neutral).

[1007] Topic identification: Identifying themes and topics based on text data.

[1008] Category sorting

[1009] Based on the analysis results, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[1010] Allocation to products

[1011] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[1012] Proposal / proposal

[1013] The server automatically generates suggestions based on the assigned text data. The generated suggestions reflect the emotions determined by the emotion engine. For example, if negative emotions are included, the server will suggest response procedures that require special attention.

[1014] Notification of proposal details

[1015] The server sends the generated proposal to the notification system, which then notifies the relevant department (user) of the proposal. The user then checks the proposal and takes specific action.

[1016] Specific examples

[1017] As a concrete example, consider the case of processing voice data from a customer saying, "The network connection is unstable." The device collects this voice data, and the server uses a speech recognition engine to convert it into text data saying, "The network connection is unstable." The server then analyzes this text data using NLP technology and an emotion engine to extract the keywords "network," "connection," and "unstable," and determines that the emotion is negative. This data is classified into the "technical support" category and assigned to the technical support department. Finally, the server generates a proposal saying, "Technical support procedures for stabilizing the network connection, and particularly, a prompt response is required," and notifies the technical support staff (user). The user then provides specific technical support to the customer based on this proposal.

[1018] In this way, this system efficiently processes voice data from customers and, by combining it with an emotion engine, provides more accurate responses, enabling quick and appropriate responses.

[1019] The processing flow will be explained below.

[1020] Step 1: Collecting audio data

[1021] The terminal receives calls and voice messages from customers in real time, and the received voice data is sent from the terminal to a server, which stores the voice data in a database.

[1022] Step 2: Convert audio data to text

[1023] The server sends the voice data stored in the database to a voice recognition engine, which converts the voice data into text data, which is then stored in the database by the server.

[1024] Step 3: Analyze the text data

[1025] The server analyzes the text data using natural language processing (NLP) technology and an emotion engine. Specifically, it uses NLP technology to perform keyword extraction and identify important keywords. It then uses the emotion engine to determine the sentiment (positive, negative, neutral) in the text data. It also performs topic identification and identifies the main themes in the text data.

[1026] Step 4: Categorize

[1027] Based on the analysis, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[1028] Step 5: Allocate to products

[1029] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[1030] Step 6: Generate proposals

[1031] The server automatically generates suggestions based on the assigned text data. The suggestions reflect the emotions determined by the emotion engine. For example, if negative emotions are included, the server will suggest procedures that require special attention.

[1032] Step 7: Notification of proposal

[1033] The server sends the generated proposal to the notification system, which notifies the person in charge (user) of the proposal.

[1034] Step 8: User Action

[1035] The user receives the proposal from the notification system and takes specific action based on the proposal. For example, in the case of technical support, the user provides technical support to the customer based on the proposal.

[1036] The above processing steps enable the system to accurately understand customer feedback and respond quickly and appropriately. By combining this system with an emotion engine, the quality of responses can be further improved.

[1037] Example 2

[1038] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1039] Conventional customer support systems had problems in efficiently collecting and analyzing voice data from customers, allocating it to the appropriate department based on its content, and quickly generating and notifying proposals. In particular, they lacked the ability to respond in a way that took customer emotions into consideration, making it difficult to improve customer satisfaction.

[1040] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1041] In this invention, the server includes a means for the terminal to collect voice data in real time or batch processing and send it to the server, a means for the server to convert the voice data into text data using a voice recognition engine, and a means for the server to analyze the text data using natural language processing technology and an emotion engine. This makes it possible to efficiently collect and analyze voice data from customers, assign it to the appropriate department based on its content, and quickly generate and notify proposals. It also enables responses that take customer emotions into consideration, which is expected to improve customer satisfaction.

[1042] "Customer Voice Data" refers to information in voice format transmitted by a Customer.

[1043] "Collection means" refers to a combination of hardware and software for capturing and recording audio data.

[1044] "Means for converting into text data" refers to technology and devices that analyze voice data and convert it into corresponding character string information.

[1045] "Means of analysis and classification" refers to the process of processing text data, analyzing its content, and classifying it into specific categories.

[1046] "Means for allocating to the corresponding department" refers to a system or algorithm for transferring classified data to the most appropriate department.

[1047] "Means for generating proposal content" refers to technology that automatically creates countermeasures and proposal content based on analyzed data.

[1048] "Means for notifying" refers to communication devices and systems for notifying the relevant departments and personnel of the content of the generated proposal.

[1049] "Real-time or batch processing" refers to a method of processing data immediately and a method of processing data in bulk.

[1050] "Speech recognition engine" refers to a software and hardware system for converting voice data into text data.

[1051] "Natural language processing technology" refers to computer technologies and algorithms for understanding, analyzing, and generating text data.

[1052] An "emotion engine" refers to software and algorithms that analyze text data and determine emotions from it.

[1053] MODE FOR CARRYING OUT THE INVENTION

[1054] This invention relates to a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies proposals. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to respond more accurately. Specific embodiments are described below.

[1055] This system consists of three components: a server, a terminal, and a user. The server processes the main data, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[1056] Audio data collection

[1057] The terminal captures voice data from the customer in real time or in batch processing using a microphone device and recording software (e.g., Audacity). The captured voice data is stored in temporary storage and later transmitted to the server.

[1058] Converting audio data to text

[1059] The server converts the received voice data into text data using a speech recognition engine (for example, Google Cloud Speech-to-Text API), and the converted text data is stored in a database.

[1060] Text data analysis

[1061] The server analyzes the text data using natural language processing technology and an emotion engine (e.g., Hume AI's emotion recognition API). The server extracts keywords using the TF-IDF algorithm and determines the user's emotion (positive, negative, neutral, etc.) using the emotion engine. It also uses natural language processing technology to identify topics.

[1062] Category sorting

[1063] The server then classifies the text data into predefined categories based on the analysis results (e.g., "technical support," "sales," "complaint," etc.) using the text classification capabilities of the Google Cloud Natural Language API.

[1064] Allocation to departments

[1065] The server assigns text data that has been classified into the appropriate category to the corresponding department, for example, data classified as "technical support" will be sent to the technical support department.

[1066] Proposal / proposal

[1067] The server automatically generates suggestions based on the assigned text data. The suggestions reflect the emotions determined by the emotion engine, and if negative emotions are included, a quick response is required.

[1068] Notification of proposal details

[1069] The server sends the generated proposal to the notification system. The notification system notifies the person in charge (user) of the proposal. The user confirms the notified proposal and takes specific action.

[1070] Specific examples

[1071] For example, consider the case where voice data from a customer saying "The network connection is unstable" is processed.

[1072] 1. The device collects and records this audio data.

[1073] 2. The server sends the audio file to the Google Cloud Speech-to-Text API and receives the text "Network connection is unstable."

[1074] 3. The server analyzes this text data using NLP technology and an emotion engine, extracts the keywords "network," "connection," and "unstable," and determines that the emotion is negative.

[1075] 4. The server categorizes this text data into the "Technical Support" category.

[1076] 5. The server sends this data to the technical support department.

[1077] 6. The server generates a suggestion: "Technical support procedures for stabilizing network connectivity and particularly rapid response is required."

[1078] 7. The server sends the generated proposal to the notification system, which then notifies the technical support staff (user). The user then provides specific technical support to the customer based on the proposal.

[1079] In this way, the system efficiently processes voice data from customers and, by combining it with an emotion engine, provides more accurate responses, enabling quick and appropriate responses.

[1080] Prompt Sentence Examples

[1081] For example, a prompt to a generative AI model might look like this:

[1082] "Convert the customer's voice data, 'The network connection is unstable,' into text data using NLP technology and a sentiment analysis engine, extract keywords, determine sentiment, and convert it into specific suggestions."

[1083] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1084] Step 1:

[1085] The terminal collects voice data in real time or in batch processing. It uses a microphone device and recording software (e.g., Audacity) to capture voice data from customers. This saves the customer's voice as an audio file, which is then sent to the server.

[1086] Specific operation: When a customer says "The network connection is unstable," the device's microphone records this, saves it as an audio file, and sends it to the server.

[1087] Input: Customer voice data

[1088] Output: Audio file

[1089] Step 2:

[1090] The server sends the received audio file to a speech recognition engine, which uses the Google Cloud Speech-to-Text API to convert the audio data into text data, which the server then stores in a database.

[1091] Specific operation: The server sends the audio file "The network connection is unstable" to the Google Cloud Speech-to-Text API, obtains the text data "The network connection is unstable", and stores it in the database.

[1092] Input: Audio file

[1093] Output: Text data

[1094] Step 3:

[1095] The server analyzes the stored text data using natural language processing technology and an emotion engine (for example, Hume AI's emotion recognition API). Specifically, it uses the TF-IDF algorithm to extract keywords, and the emotion engine determines the user's emotion (positive, negative, neutral, etc.). It also uses natural language processing technology to identify topics.

[1096] Specific operation: The server analyzes the text data "Your network connection is unstable," extracts the keywords "network," "connection," and "unstable," and determines that the content contains negative sentiment.

[1097] Input: Text data

[1098] Output: Keywords, sentiment analysis, identified topics

[1099] Step 4:

[1100] The server classifies the text data into predefined categories based on the analysis results, such as "technical support," "sales," and "complaints," using the Google Cloud Natural Language API.

[1101] Specific operation: The server classifies the text data "Network connection is unstable" into the "Technical Support" category.

[1102] Input: Keywords, sentiment analysis results, identified topics

[1103] Output: Categorized data

[1104] Step 5:

[1105] The server assigns text data that has been classified into the appropriate category to the corresponding department, for example, data classified as "technical support" is sent to the technical support department.

[1106] What happens: The server sends data categorized under "Tech Support" to the technical support department.

[1107] Input: Categorical data

[1108] Output: Data assigned to departments

[1109] Step 6:

[1110] The server automatically generates suggestions based on the assigned text data, taking into account the emotions (e.g., negative) determined by the emotion engine.

[1111] What happens: The server generates a suggestion that states, "Technical support procedures for stabilizing network connectivity, and particularly prompt action, are needed."

[1112] Input: Data assigned to department, emotion judgment result

[1113] Output: Proposal

[1114] Step 7:

[1115] The server sends the generated proposal to the notification system. The notification system notifies the person in charge (user) of the proposal. The user confirms the notified proposal and takes specific action.

[1116] Specific operation: The server sends the proposal to the notification system, which notifies the technical support staff (user). The user provides specific technical support based on the proposal.

[1117] Input: Proposal

[1118] Output: Proposal notified

[1119] (Application example 2)

[1120] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1121] Conventional customer service systems have had the problem of being unable to efficiently collect customer feedback, quickly analyze and classify it, and make appropriate proposals. In particular, when dealing with customers in physical stores, real-time feedback is required, but conventional systems were inadequate in this regard.

[1122] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data, means for assigning the classified data to the corresponding department, means for generating proposals based on the assigned data, means for notifying the generated proposals, and smartphone application means for responding to customers in real time at the store. This makes it possible to collect and analyze customer feedback in real time at a physical store and quickly make appropriate proposals.

[1123] "Customer voice data" refers to voice information of words and phrases spoken by a customer.

[1124] "Means of collection" refers to devices or systems for capturing audio data, such as microphones or voice recording applications.

[1125] "Means for converting into text data" refers to a voice recognition engine or software for converting voice data into text information.

[1126] "Means for analyzing and classifying" refers to natural language processing techniques and algorithms used to analyze the converted text data and classify it into specific categories.

[1127] The "means for allocating to the corresponding department" refers to a system, such as a database management system, that automatically sorts the analyzed and classified data to the appropriate business department.

[1128] "Means for generating recommendations" refers to software algorithms that automatically generate solutions or advice based on the assigned data.

[1129] "Means of notification" refers to messaging and alert systems that notify relevant department staff of generated suggestions and countermeasures.

[1130] "Customer service in stores" refers to the business activities of responding quickly to customer requests and questions in physical stores.

[1131] "Smartphone application means" refers to dedicated application software that allows users to respond to customers via their smartphones.

[1132] The system for implementing this invention is composed of three entities: a server, a terminal, and a user. The specific roles and processing contents of each entity are explained below.

[1133] Audio data collection and text conversion

[1134] The terminal uses a dedicated application installed on a smartphone to handle customer inquiries in real time within the store. The smartphone's microphone is used to collect customer voice data. This voice data is then sent to a server, where it is converted into text data using a speech recognition engine (for example, Google's speech recognition API). The converted text data is then stored on the server.

[1135] Analysis and classification of text data

[1136] The server then analyzes the converted text data using natural language processing (NLP) techniques, such as Hugging Face's BERT model, to extract keywords, analyze sentiment, and identify topics from the text data. It also uses an emotion engine to determine the user's sentiment (e.g., positive, negative, neutral) based on the content of the text data. Based on the analysis results, the text data is classified into predefined categories (e.g., technical support, sales, complaints).

[1137] Allocating data and generating proposals

[1138] The server automatically assigns the classified text data to the appropriate department. For example, data classified as "technical support" is sent to the technical support department. Next, a suggestion is automatically generated. The generated suggestion reflects the emotion determined by the emotion engine. If a negative emotion is included, a response procedure requiring special attention is suggested.

[1139] Notification and implementation of proposals

[1140] The server sends the generated proposal to a notification system, which notifies staff in the relevant departments. Notification methods include push notifications via smartphone applications and internal messaging systems. The users (staff) then check the notified proposal and implement specific measures. This enables quick and appropriate responses that reflect customer feedback in real time.

[1141] Adding specific examples

[1142] As a concrete example, consider the case where a customer in a store says, "I don't know where the product is." This voice data is collected by a smartphone app and converted into text data, "I don't know where the product is," using a voice recognition engine. The server then analyzes this text data using NLP technology and an emotion engine, extracting the keywords "product," "location," and "don't know," and determining that the emotion is negative. This data is classified into the "store support" category and assigned to the relevant department. Finally, the server generates a suggestion, "You need guidance to the product location and prompt customer service," and notifies the staff.

[1143] Prompt Sentence Examples

[1144] Voice data is collected and converted into text data. The converted text data is analyzed using NLP technology to extract keywords, analyze sentiment, and identify topics. Based on the results, appropriate suggestions are generated and notified to the relevant departments.

[1145] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1146] Step 1: Collecting audio data

[1147] The device (smartphone) collects the customer's voice data through the application using a microphone. The input is the customer's voice data, and the output is the collected voice file. This voice file is sent directly to the server.

[1148] Step 2: Convert audio data to text

[1149] The server receives the transmitted audio file and converts it into text data using a speech recognition engine. The input is the audio file, and the output is the converted text data. Specifically, the server analyzes the audio data using Google's speech recognition API and generates the corresponding text data.

[1150] Step 3: Save the text data

[1151] The server stores the converted text data in a database. The input of this process is the text data obtained in the previous step, and the output is the stored text data. A database management system is used to efficiently store the text data.

[1152] Step 4: Analyze and classify text data

[1153] The server analyzes the stored text data using natural language processing (NLP) technology and a sentiment analysis engine. Specifically, it uses the Hugging Face BERT model to extract keywords, analyze sentiment, and identify topics from the text data. The input is the text data, and the output is the extracted keywords, sentiment classification, and topic identification results.

[1154] Step 5: Allocate data

[1155] The server automatically assigns data to the appropriate department based on the analyzed and classified data. The input is keywords, sentiment classification, and topic identification results, and the output is department assignment information. This is done based on pre-defined categories.

[1156] Step 6: Generate proposals

[1157] The server generates appropriate proposals based on the assigned data. The input is the assignment information to the corresponding department, and the output is the generated proposal. The proposals are automatically created using an algorithm and include appropriate countermeasures.

[1158] Step 7: Notification of proposal

[1159] The server sends the generated proposal to a notification system, which notifies staff in the relevant departments. The input is the proposal, and the output is the notified information. Notification methods include push notifications on smartphones and internal messaging systems.

[1160] Step 8: Implementing customer interactions

[1161] The user (staff member) checks the notified proposal and takes specific action. The input is the notified proposal, and the output is the result of the action taken. This enables quick and appropriate action that reflects customer feedback in real time.

[1162] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1163] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1164] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1165] [Fourth embodiment]

[1166] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1167] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1168] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1169] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1170] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1171] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1172] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1173] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1174] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1175] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1176] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1177] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1178] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1179] This invention is a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies the customer of proposals. A specific embodiment of this system will be described.

[1180] System Overview

[1181] This system is mainly composed of three components: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[1182] Audio data collection

[1183] The terminal captures voice data from the customer in real time or in batch processing, and the captured voice data is sent to the server and stored in a database.

[1184] Converting audio data to text

[1185] The server sends the saved voice data to a voice recognition engine, which converts the voice data into text data and saves the text data back in the database.

[1186] Text data analysis

[1187] The server analyzes the text data using natural language processing (NLP) techniques, which specifically perform the following operations:

[1188] Keyword extraction: Using specific algorithms to extract key keywords from text data.

[1189] Sentiment analysis: Determine the sentiment (positive, negative, neutral) from the content of text data.

[1190] Topic identification: Identifying themes and topics based on text data.

[1191] Category sorting

[1192] The server analyzes the text data and classifies it into predefined categories, such as "technical support," "sales," and "complaints."

[1193] Allocation to products

[1194] The server assigns categorized text data to the appropriate department, for example, data categorized in the "Tech Support" category is sent to the Tech Support department.

[1195] Proposal / proposal

[1196] The server automatically generates proposals based on the assigned data. The generated proposals are notified to the person in charge (user) of the corresponding department via the notification system. The user then checks the notified proposals and takes specific action.

[1197] Specific examples

[1198] As a concrete example, consider the case of processing voice data from a customer saying, "The network connection is unstable." The device collects this voice data, and the server uses a speech recognition engine to convert it into text data saying, "The network connection is unstable." The server then analyzes this text data using NLP technology to extract the keywords "network," "connection," and "unstable," and determines that the sentiment is negative. This data is classified into the "technical support" category and assigned to the technical support department. Finally, the server generates a proposal, "technical support procedures for stabilizing the network connection," and notifies the technical support representative (user). The user then provides specific technical support to the customer based on this proposal.

[1199] In this way, this system efficiently processes voice data from customers and enables quick and appropriate responses, thereby improving customer satisfaction and operational efficiency.

[1200] The processing flow will be explained below.

[1201] Step 1: Collecting audio data

[1202] The terminal receives calls and voice messages from customers. The received voice data is sent to the server in real time or in batch processing. The server stores this voice data in a database.

[1203] Step 2: Convert audio data to text

[1204] The server sends the voice data stored in the database to a voice recognition engine, which converts the voice data into text data, which is then stored in the database by the server.

[1205] Step 3: Analyze the text data

[1206] The server analyzes the text data using natural language processing (NLP) techniques. The server applies keyword extraction algorithms to extract important keywords. Furthermore, the server performs sentiment analysis to determine the sentiment in the text (e.g., positive, negative, neutral). Finally, the server performs topic identification to recognize the subject matter of the text data.

[1207] Step 4: Categorize

[1208] Based on the analysis, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[1209] Step 5: Allocate to products

[1210] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[1211] Step 6: Generate proposals

[1212] The server automatically generates suggestions based on the assigned text data. For example, if the data is related to technical support, it generates technical procedures for stabilizing network connections.

[1213] Step 7: Notification of proposal

[1214] The server sends the generated proposal to the notification system, which then notifies the relevant department (user) of the proposal. The user then checks the proposal and takes specific action.

[1215] The above processing steps enable accurate understanding of customer feedback and prompt and appropriate response.

[1216] Example 1

[1217] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1218] Conventional systems have difficulty efficiently collecting voice data from customers, appropriately converting it into text data, analyzing it, classifying it, and generating and notifying customers of proposals. In particular, the lack of keyword extraction, sentiment analysis, or topic identification during voice data analysis means that appropriate responses tend to be delayed. Furthermore, because proposals are not automatically generated using generative AI models, there is room for improvement in the accuracy and speed of responses. There is a need for a system that can resolve these issues and enable faster and more accurate responses.

[1219] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1220] In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data by extracting keywords, analyzing sentiment, and identifying topics using natural language processing technology, means for assigning the classified data to an appropriate department based on predefined categories, means for generating proposals using a generative AI model based on the assigned data, and means for notifying the generated proposals. This enables efficient and accurate processing of voice data from customers, enabling prompt and accurate responses.

[1221] "Customer voice data" refers to voice information transmitted by customers, including customer questions, complaints, requests, etc.

[1222] "Means for collection" refers to the device or software used to capture voice data from customers and input it into the system.

[1223] "Means for converting voice data into text data" refers to a device or software that converts voice data into text data using voice recognition technology.

[1224] "Means for analyzing text data" refers to a device or software that uses natural language processing techniques to analyze acquired text data.

[1225] "Means for classifying" refers to a device or software that appropriately classifies analyzed text data into predefined categories.

[1226] A "predefined category" is a category that is set in advance for classifying data, and examples include "technical support," "sales," and "complaints."

[1227] "Means for allocating to appropriate departments" refers to devices or software that transmit or distribute classified data to the departments corresponding to each category.

[1228] "Generative AI model" refers to an artificial intelligence model that uses deep learning technology to generate suggestions based on specific input data.

[1229] "Means for generating proposals" refers to devices or software that use generative AI models to automatically create proposals and solutions based on customer feedback.

[1230] "Means for notifying" refers to a device or software for communicating the generated proposal to an appropriate recipient.

[1231] A "prompt sentence" refers to text data that is input to a generative AI model to generate a specific output.

[1232] The present invention is a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies the customer of proposals. Specific embodiments are described below.

[1233] This system is mainly composed of three entities: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[1234] Audio data collection

[1235] The terminal acquires voice data from the customer through a user-operated interface. Specifically, the terminal is equipped with a microphone and recording software (e.g., Audacity) that is used to collect voice data in real time or in batches. The collected voice data is encrypted and transmitted to a server using a secure communication protocol (e.g., HTTPS).

[1236] Converting audio data to text

[1237] The server stores the received voice data in a database. The server then sends the stored voice data to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice data into text data, which is then stored in the database again.

[1238] Text data analysis

[1239] The server analyzes the stored text data using natural language processing (NLP) techniques, specifically performing the following operations:

[1240] Keyword extraction: Uses the TF-IDF algorithm to extract key keywords from text data.

[1241] Sentiment analysis: Use a sentiment analysis API (e.g., IBM Watson Natural Language Understanding) to classify the sentiment of text data as positive, negative, or neutral.

[1242] Topic Identification: Identify topics within text data using pre-trained topic models (e.g., Latent Dirichlet Allocation).

[1243] Category sorting

[1244] Based on the results of the text data analysis, the server classifies the data into predefined categories (e.g., "technical support," "sales," "complaints") using the Naive Bayes classifier as the classification algorithm.

[1245] Allocation to products

[1246] The server assigns the classified text data to the appropriate department, using a department collaboration system (e.g., Slack or Microsoft Teams).

[1247] Proposal / proposal

[1248] The server automatically generates proposals using a generative AI model (e.g., OpenAI's GPT-3) based on the assigned data. The generated proposals are notified to the person in charge via a notification system (e.g., email or SMS gateway). The user receives the notification and takes specific action.

[1249] Specific examples

[1250] As a specific example, consider a case where voice data from a customer saying "The network connection is unstable" is collected by a terminal.

[1251] 1. The device collects the customer's voice message saying "The network connection is unstable," encrypts it, and sends it to the server.

[1252] 2. The server stores the voice data in a database and uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert it into text data such as "The network connection is unstable."

[1253] 3. The server analyzes this text data using natural language processing technology (e.g., IBM Watson Natural Language Understanding), extracts the keywords "network," "connection," and "unstable," and determines that the sentiment is negative.

[1254] 4. This data is classified under the category "Technical Support" and assigned to the Technical Support department.

[1255] 5. The server generates a suggestion by inputting the following prompt into the generative AI model (e.g., OpenAI's GPT-3): "The network connection is unstable. Please suggest technical support steps to resolve this."

[1256] 6. The generated proposal, "Technical support procedure for stabilizing network connections," is notified to the technical support staff. The technical support staff (user) provides specific technical support to the customer based on this proposal.

[1257] In this way, the system efficiently processes voice data from customers, enabling quick and accurate responses.

[1258] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1259] Step 1: Collecting audio data

[1260] Input: Customer voice input

[1261] Output: Audio data saved on the device (file format: .wav, etc.)

[1262] Specific operation:

[1263] The terminal utilizes a microphone and recording software (e.g., Audacity) to record audio in real time through the customer interface.

[1264] A customer says, "My network connection is unstable."

[1265] The device captures the audio and temporarily stores it in .wav format.

[1266] The saved audio data is encrypted and sent to the server.

[1267] Step 2: Convert audio data to text

[1268] Input: Collected encrypted audio data

[1269] Output: Text data

[1270] Specific operation:

[1271] The server stores the received voice data in a database.

[1272] The saved voice data is sent to a speech recognition engine (e.g., Google Cloud Speech-to-Text API) and converted into text data.

[1273] The speech recognition engine returns text data saying "Network connection is unstable."

[1274] The server stores the converted text data in a database.

[1275] Step 3: Analyze the text data

[1276] Input: Text data

[1277] Output: Analysis results (keywords, sentiment, topics)

[1278] Specific operation:

[1279] The server analyzes the stored text data using natural language processing (NLP) technology.

[1280] Keyword extraction: The keywords “network”, “connection”, and “unstable” are extracted using the TF-IDF algorithm.

[1281] Sentiment analysis: Use a sentiment analysis API (e.g., IBM Watson Natural Language Understanding) to determine whether the text data is negative.

[1282] Topic Identification: Identify the topic "stabilizing network connectivity" using a pre-trained topic model (e.g., Latent Dirichlet Allocation).

[1283] Step 4: Categorize

[1284] Input: Analysis results (keywords, sentiment, topics)

[1285] Output: Classified category information

[1286] Specific operation:

[1287] Based on the analysis, the server classifies the data into predefined categories (e.g., "technical support").

[1288] The Naive Bayes classifier is used as the classification algorithm.

[1289] Information categorized under "Technical Support" is generated.

[1290] Step 5: Assign to departments

[1291] Input: Classified category information

[1292] Output: Data assigned to departments

[1293] Specific operation:

[1294] The server assigns data categorized under the "Technical Support" category to the Technical Support department.

[1295] Send data to the technical support department in real time using a cross-departmental system (e.g., Slack or Microsoft Teams).

[1296] Step 6: Proposal and project creation

[1297] Input: Data assigned to department

[1298] Output: Generated proposals

[1299] Specific operation:

[1300] The server generates suggestions using a generative AI model (e.g., OpenAI's GPT-3) based on the assigned data.

[1301] Enter the prompt sentence into the generative AI model: "The network connection is unstable. Please suggest technical support steps to resolve it."

[1302] The generative AI model generates "technical support procedures for stabilizing network connections," which are output as suggestions.

[1303] Step 7: Notification

[1304] Input: Generated proposal

[1305] Output: The suggestion sent to the user

[1306] Specific operation:

[1307] The server notifies the generated proposal to the technical support personnel via a notification system (e.g., email or SMS gateway).

[1308] The user receives a notification and takes specific action.

[1309] (Application example 1)

[1310] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1311] Current food delivery systems lack the means to quickly and accurately collect and analyze customer feedback. In particular, there is no system that can efficiently process voice feedback, which results in issues such as time spent improving customer satisfaction and resolving problems. Furthermore, conventional systems have difficulty accurately understanding customer sentiment and the content of feedback and quickly directing the appropriate department to respond.

[1312] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1313] In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data, means for assigning the classified data to a corresponding department, means for generating proposals based on the assigned data, means for notifying the generated proposals, means for extracting keywords from customer feedback using the voice data, and means for performing sentiment analysis using natural language processing technology to classify categories based on the keywords. This makes it possible to efficiently collect and analyze customer voice feedback and quickly and appropriately notify the corresponding department of proposals.

[1314] "Customer voice data" refers to voice information generated by customers, including feedback and inquiries about services and products.

[1315] "Means" refers to devices, methods, techniques or processes used to achieve a particular purpose.

[1316] "Text data" refers to information converted from audio into a string of characters, such as a collection of sentences or words.

[1317] "Keyword extraction" refers to the process of selecting key words and phrases from text data.

[1318] "Sentiment analysis" is the process of determining the writer's emotions from the content of text data, which are classified as positive, negative, neutral, etc.

[1319] "Natural language processing technology" refers to the technology that enables computers to understand, interpret, and generate human language, and includes keyword extraction and sentiment analysis as part of this.

[1320] A "department" refers to a specialized team responsible for a specific task or function within an organization.

[1321] "Proposals" refer to solutions or countermeasures that are automatically generated based on the analyzed data.

[1322] "Real-time" refers to data collection and processing occurring immediately, with minimal delay.

[1323] "Sentiment analysis using natural language processing technology" refers to the technology of using a computer to analyze emotions from text data, and is a part of natural language processing technology.

[1324] A "system" refers to an overall structure in which multiple components and means work together to perform a specific function.

[1325] This invention provides a system for food delivery services that efficiently collects and analyzes customer voice feedback and promptly notifies the relevant department of the proposed solution. This system consists of three entities: a server, a terminal, and a user.

[1326] System Overview

[1327] server

[1328] The server handles the central data processing: converting voice data to text, analyzing text data, allocating data to products, generating recommendations, and notifications.

[1329] Terminal

[1330] The device collects voice data from customers in real time. The data is picked up through a microphone and sent to a server. The device also serves as an interface for customers and guides them through the feedback collection process.

[1331] User

[1332] The user checks the proposals sent by the system and implements specific measures. The user refers to the person in charge of the relevant department, such as a technical support staff member or a manager.

[1333] Audio data collection

[1334] The device collects voice data from the microphone using the speech_recognition library. The voice information spoken by the customer is captured in real time and sent to the server. The collected voice data is converted to text data using the recognizer.recognize_google method.

[1335] Text data analysis

[1336] The server analyzes the acquired text data using natural language processing technology, specifically by performing the following operations:

[1337] Keyword Extraction: Use TextBlob to extract key words and phrases from text data.

[1338] Sentiment Analysis: Use TextBlob to classify the sentiment of text (positive, negative, neutral).

[1339] Category classification: Uses the transformers library pipeline to determine which category the text belongs to.

[1340] Product allocation and proposal generation

[1341] Based on the analysis results, the server assigns the data to the appropriate department and generates a recommendation for the corresponding department. For example, if the request is "feedback on an order," it will be assigned to the order processing department and a recommendation such as "Please check the order."

[1342] Notification of proposal details

[1343] The generated proposals are automatically notified to the relevant departments. The proposals are provided instantly through a notification system (pseudo implementation).

[1344] Specific examples

[1345] Below is an example of customer feedback:

[1346] "I couldn't specify a delivery time."

[1347] "The food tastes different than usual"

[1348] "The app is hard to understand"

[1349] For these specific examples, gather audio feedback using the following example prompts:

[1350] Example prompt sentence:

[1351] "Please provide your feedback by voice. For example, 'I can't confirm my order!' or 'The food tastes different.'"

[1352] Hardware and software used

[1353] Hardware: Microphone (for audio input)

[1354] Software: Python, speech_recognition, transformers, TextBlob

[1355] Data processing and calculation

[1356] The server converts the voice data into text and performs keyword extraction, sentiment analysis, and categorization, allowing it to properly categorize customer feedback and quickly notify the relevant department with suggestions.

[1357] In this way, the present invention provides a system that improves customer satisfaction in food delivery services and also improves operational efficiency.

[1358] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1359] Step 1:

[1360] The terminal collects voice data. Specifically, it obtains voice feedback from the customer through a microphone and collects it in real time using a speech recognition engine (speech_recognition). The input is the customer's voice, and the output is voice data.

[1361] Step 2:

[1362] The terminal transmits the collected voice data to the server. The input is the voice data, and the output is the voice data transmitted to the server.

[1363] Step 3:

[1364] The server converts the received voice data into text data. Specifically, it converts the voice data into text using the recognizer.recognize_google method. The input is voice data, and the output is text data.

[1365] Step 4:

[1366] The server analyzes the converted text data. Specifically, it extracts keywords from the text data (using TextBlob), performs sentiment analysis (also using TextBlob), and classifies it into appropriate categories (using a pipeline from the transformers library). The input is text data, and the output is keywords, sentiment classification, and categorized data.

[1367] Step 5:

[1368] The server assigns data to the corresponding department based on the analysis results. The input is keywords, sentiment classification, and categorized data, and the output is the assignment of data to the corresponding department.

[1369] Step 6:

[1370] The server generates proposals based on the assigned data. Specifically, it automatically proposes predefined countermeasures based on the analyzed categories. The input is the data assigned to the corresponding department, and the output is the proposals.

[1371] Step 7:

[1372] The server notifies the person in charge of the corresponding department of the content of the generated proposal via the notification system. The input is the content of the proposal, and the output is the notification to the person in charge.

[1373] Step 8:

[1374] The user (person in charge) implements specific measures based on the proposal. The input is the proposal, and the output is the implementation of specific measures.

[1375] This series of steps enables efficient processing of voice feedback from customers and provides prompt and appropriate responses.

[1376] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1377] The present invention relates to a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies proposals. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to respond more accurately. Specific embodiments of this system will be described below.

[1378] System Overview

[1379] This system consists of three components: a server, a terminal, and a user. The server performs central data processing, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[1380] Audio data collection

[1381] The terminal captures voice data from the customer in real time or in batch processing, and the captured voice data is sent to the server and stored in a database.

[1382] Converting audio data to text

[1383] The server sends the saved voice data to a voice recognition engine, which converts the voice data into text data, which is then saved in a database by the server.

[1384] Text data analysis

[1385] The server analyzes the text data using natural language processing (NLP) techniques and an emotion engine. Specifically, this process involves the following operations:

[1386] Keyword extraction: Using specific algorithms to extract key keywords from text data.

[1387] Sentiment analysis: Using a sentiment engine to determine user sentiment from the content of text data (e.g., positive, negative, neutral).

[1388] Topic identification: Identifying themes and topics based on text data.

[1389] Category sorting

[1390] Based on the analysis results, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[1391] Allocation to products

[1392] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[1393] Proposal / proposal

[1394] The server automatically generates suggestions based on the assigned text data. The generated suggestions reflect the emotions determined by the emotion engine. For example, if negative emotions are included, the server will suggest response procedures that require special attention.

[1395] Notification of proposal details

[1396] The server sends the generated proposal to the notification system, which then notifies the relevant department (user) of the proposal. The user then checks the proposal and takes specific action.

[1397] Specific examples

[1398] As a concrete example, consider the case of processing voice data from a customer saying, "The network connection is unstable." The device collects this voice data, and the server uses a speech recognition engine to convert it into text data saying, "The network connection is unstable." The server then analyzes this text data using NLP technology and an emotion engine to extract the keywords "network," "connection," and "unstable," and determines that the emotion is negative. This data is classified into the "technical support" category and assigned to the technical support department. Finally, the server generates a proposal saying, "Technical support procedures for stabilizing the network connection, and particularly, a prompt response is required," and notifies the technical support staff (user). The user then provides specific technical support to the customer based on this proposal.

[1399] In this way, this system efficiently processes voice data from customers and, by combining it with an emotion engine, provides more accurate responses, enabling quick and appropriate responses.

[1400] The processing flow will be explained below.

[1401] Step 1: Collecting audio data

[1402] The terminal receives calls and voice messages from customers in real time, and the received voice data is sent from the terminal to a server, which stores the voice data in a database.

[1403] Step 2: Convert audio data to text

[1404] The server sends the voice data stored in the database to a voice recognition engine, which converts the voice data into text data, which is then stored in the database by the server.

[1405] Step 3: Analyze the text data

[1406] The server analyzes the text data using natural language processing (NLP) technology and an emotion engine. Specifically, it uses NLP technology to perform keyword extraction and identify important keywords. It then uses the emotion engine to determine the sentiment (positive, negative, neutral) in the text data. It also performs topic identification and identifies the main themes in the text data.

[1407] Step 4: Categorize

[1408] Based on the analysis, the server classifies the text data into predefined categories, such as "technical support," "sales," and "complaints."

[1409] Step 5: Allocate to products

[1410] The server assigns each category of text data to the appropriate internal department. For example, data categorized as "technical support" is sent to the technical support department.

[1411] Step 6: Generate proposals

[1412] The server automatically generates suggestions based on the assigned text data. The suggestions reflect the emotions determined by the emotion engine. For example, if negative emotions are included, the server will suggest procedures that require special attention.

[1413] Step 7: Notification of proposal

[1414] The server sends the generated proposal to the notification system, which notifies the person in charge (user) of the proposal.

[1415] Step 8: User Action

[1416] The user receives the proposal from the notification system and takes specific action based on the proposal. For example, in the case of technical support, the user provides technical support to the customer based on the proposal.

[1417] The above processing steps enable the system to accurately understand customer feedback and respond quickly and appropriately. By combining this system with an emotion engine, the quality of responses can be further improved.

[1418] Example 2

[1419] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1420] Conventional customer support systems had problems in efficiently collecting and analyzing voice data from customers, allocating it to the appropriate department based on its content, and quickly generating and notifying proposals. In particular, they lacked the ability to respond in a way that took customer emotions into consideration, making it difficult to improve customer satisfaction.

[1421] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1422] In this invention, the server includes a means for the terminal to collect voice data in real time or batch processing and send it to the server, a means for the server to convert the voice data into text data using a voice recognition engine, and a means for the server to analyze the text data using natural language processing technology and an emotion engine. This makes it possible to efficiently collect and analyze voice data from customers, assign it to the appropriate department based on its content, and quickly generate and notify proposals. It also enables responses that take customer emotions into consideration, which is expected to improve customer satisfaction.

[1423] "Customer Voice Data" refers to information in voice format transmitted by a Customer.

[1424] "Collection means" refers to a combination of hardware and software for capturing and recording audio data.

[1425] "Means for converting into text data" refers to technology and devices that analyze voice data and convert it into corresponding character string information.

[1426] "Means of analysis and classification" refers to the process of processing text data, analyzing its content, and classifying it into specific categories.

[1427] "Means for allocating to the corresponding department" refers to a system or algorithm for transferring classified data to the most appropriate department.

[1428] "Means for generating proposal content" refers to technology that automatically creates countermeasures and proposal content based on analyzed data.

[1429] "Means for notifying" refers to communication devices and systems for notifying the relevant departments and personnel of the content of the generated proposal.

[1430] "Real-time or batch processing" refers to a method of processing data immediately and a method of processing data in bulk.

[1431] "Speech recognition engine" refers to a software and hardware system for converting voice data into text data.

[1432] "Natural language processing technology" refers to computer technologies and algorithms for understanding, analyzing, and generating text data.

[1433] An "emotion engine" refers to software and algorithms that analyze text data and determine emotions from it.

[1434] MODE FOR CARRYING OUT THE INVENTION

[1435] This invention relates to a system that efficiently collects customer voice data, converts the data into text data, analyzes and classifies it, assigns it to the appropriate department, and then generates and notifies proposals. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to respond more accurately. Specific embodiments are described below.

[1436] This system consists of three components: a server, a terminal, and a user. The server processes the main data, and the terminal provides an interface with the user. The user is responsible for dealing with customers and implementing their proposals.

[1437] Audio data collection

[1438] The terminal captures voice data from the customer in real time or in batch processing using a microphone device and recording software (e.g., Audacity). The captured voice data is stored in temporary storage and later transmitted to the server.

[1439] Converting audio data to text

[1440] The server converts the received voice data into text data using a speech recognition engine (for example, Google Cloud Speech-to-Text API), and the converted text data is stored in a database.

[1441] Text data analysis

[1442] The server analyzes the text data using natural language processing technology and an emotion engine (e.g., Hume AI's emotion recognition API). The server extracts keywords using the TF-IDF algorithm and determines the user's emotion (positive, negative, neutral, etc.) using the emotion engine. It also uses natural language processing technology to identify topics.

[1443] Category sorting

[1444] The server then classifies the text data into predefined categories based on the analysis results (e.g., "technical support," "sales," "complaint," etc.) using the text classification capabilities of the Google Cloud Natural Language API.

[1445] Allocation to departments

[1446] The server assigns text data that has been classified into the appropriate category to the corresponding department, for example, data classified as "technical support" will be sent to the technical support department.

[1447] Proposal / proposal

[1448] The server automatically generates suggestions based on the assigned text data. The suggestions reflect the emotions determined by the emotion engine, and if negative emotions are included, a quick response is required.

[1449] Notification of proposal details

[1450] The server sends the generated proposal to the notification system. The notification system notifies the person in charge (user) of the proposal. The user confirms the notified proposal and takes specific action.

[1451] Specific examples

[1452] For example, consider the case where voice data from a customer saying "The network connection is unstable" is processed.

[1453] 1. The device collects and records this audio data.

[1454] 2. The server sends the audio file to the Google Cloud Speech-to-Text API and receives the text "Network connection is unstable."

[1455] 3. The server analyzes this text data using NLP technology and an emotion engine, extracts the keywords "network," "connection," and "unstable," and determines that the emotion is negative.

[1456] 4. The server categorizes this text data into the "Technical Support" category.

[1457] 5. The server sends this data to the technical support department.

[1458] 6. The server generates a suggestion: "Technical support procedures for stabilizing network connectivity and particularly rapid response is required."

[1459] 7. The server sends the generated proposal to the notification system, which then notifies the technical support staff (user). The user then provides specific technical support to the customer based on the proposal.

[1460] In this way, the system efficiently processes voice data from customers and, by combining it with an emotion engine, provides more accurate responses, enabling quick and appropriate responses.

[1461] Prompt Sentence Examples

[1462] For example, a prompt to a generative AI model might look like this:

[1463] "Convert the customer's voice data, 'The network connection is unstable,' into text data using NLP technology and a sentiment analysis engine, extract keywords, determine sentiment, and convert it into specific suggestions."

[1464] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1465] Step 1:

[1466] The terminal collects voice data in real time or in batch processing. It uses a microphone device and recording software (e.g., Audacity) to capture voice data from customers. This saves the customer's voice as an audio file, which is then sent to the server.

[1467] Specific operation: When a customer says "The network connection is unstable," the device's microphone records this, saves it as an audio file, and sends it to the server.

[1468] Input: Customer voice data

[1469] Output: Audio file

[1470] Step 2:

[1471] The server sends the received audio file to a speech recognition engine, which uses the Google Cloud Speech-to-Text API to convert the audio data into text data, which the server then stores in a database.

[1472] Specific operation: The server sends the audio file "The network connection is unstable" to the Google Cloud Speech-to-Text API, obtains the text data "The network connection is unstable", and stores it in the database.

[1473] Input: Audio file

[1474] Output: Text data

[1475] Step 3:

[1476] The server analyzes the stored text data using natural language processing technology and an emotion engine (for example, Hume AI's emotion recognition API). Specifically, it uses the TF-IDF algorithm to extract keywords, and the emotion engine determines the user's emotion (positive, negative, neutral, etc.). It also uses natural language processing technology to identify topics.

[1477] Specific operation: The server analyzes the text data "Your network connection is unstable," extracts the keywords "network," "connection," and "unstable," and determines that the content contains negative sentiment.

[1478] Input: Text data

[1479] Output: Keywords, sentiment analysis, identified topics

[1480] Step 4:

[1481] The server classifies the text data into predefined categories based on the analysis results, such as "technical support," "sales," and "complaints," using the Google Cloud Natural Language API.

[1482] Specific operation: The server classifies the text data "Network connection is unstable" into the "Technical Support" category.

[1483] Input: Keywords, sentiment analysis results, identified topics

[1484] Output: Categorized data

[1485] Step 5:

[1486] The server assigns text data that has been classified into the appropriate category to the corresponding department, for example, data classified as "technical support" is sent to the technical support department.

[1487] What happens: The server sends data categorized under "Tech Support" to the technical support department.

[1488] Input: Categorical data

[1489] Output: Data assigned to departments

[1490] Step 6:

[1491] The server automatically generates suggestions based on the assigned text data, taking into account the emotions (e.g., negative) determined by the emotion engine.

[1492] What happens: The server generates a suggestion that states, "Technical support procedures for stabilizing network connectivity, and particularly prompt action, are needed."

[1493] Input: Data assigned to department, emotion judgment result

[1494] Output: Proposal

[1495] Step 7:

[1496] The server sends the generated proposal to the notification system. The notification system notifies the person in charge (user) of the proposal. The user confirms the notified proposal and takes specific action.

[1497] Specific operation: The server sends the proposal to the notification system, which notifies the technical support staff (user). The user provides specific technical support based on the proposal.

[1498] Input: Proposal

[1499] Output: Proposal notified

[1500] (Application example 2)

[1501] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1502] Conventional customer service systems have had the problem of being unable to efficiently collect customer feedback, quickly analyze and classify it, and make appropriate proposals. In particular, when dealing with customers in physical stores, real-time feedback is required, but conventional systems were inadequate in this regard.

[1503] The identification process by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting customer voice data, means for converting the voice data into text data, means for analyzing and classifying the text data, means for assigning the classified data to the corresponding department, means for generating proposals based on the assigned data, means for notifying the generated proposals, and smartphone application means for responding to customers in real time at the store. This makes it possible to collect and analyze customer feedback in real time at a physical store and quickly make appropriate proposals.

[1504] "Customer voice data" refers to voice information of words and phrases spoken by a customer.

[1505] "Means of collection" refers to devices or systems for capturing audio data, such as microphones or voice recording applications.

[1506] "Means for converting into text data" refers to a voice recognition engine or software for converting voice data into text information.

[1507] "Means for analyzing and classifying" refers to natural language processing techniques and algorithms used to analyze the converted text data and classify it into specific categories.

[1508] The "means for allocating to the corresponding department" refers to a system, such as a database management system, that automatically sorts the analyzed and classified data to the appropriate business department.

[1509] "Means for generating recommendations" refers to software algorithms that automatically generate solutions or advice based on the assigned data.

[1510] "Means of notification" refers to messaging and alert systems that notify relevant department staff of generated suggestions and countermeasures.

[1511] "Customer service in stores" refers to the business activities of responding quickly to customer requests and questions in physical stores.

[1512] "Smartphone application means" refers to dedicated application software that allows users to respond to customers via their smartphones.

[1513] The system for implementing this invention is composed of three entities: a server, a terminal, and a user. The specific roles and processing contents of each entity are explained below.

[1514] Audio data collection and text conversion

[1515] The terminal uses a dedicated application installed on a smartphone to handle customer inquiries in real time within the store. The smartphone's microphone is used to collect customer voice data. This voice data is then sent to a server, where it is converted into text data using a speech recognition engine (for example, Google's speech recognition API). The converted text data is then stored on the server.

[1516] Analysis and classification of text data

[1517] The server then analyzes the converted text data using natural language processing (NLP) techniques, such as Hugging Face's BERT model, to extract keywords, analyze sentiment, and identify topics from the text data. It also uses an emotion engine to determine the user's sentiment (e.g., positive, negative, neutral) based on the content of the text data. Based on the analysis results, the text data is classified into predefined categories (e.g., technical support, sales, complaints).

[1518] Allocating data and generating proposals

[1519] The server automatically assigns the classified text data to the appropriate department. For example, data classified as "technical support" is sent to the technical support department. Next, a suggestion is automatically generated. The generated suggestion reflects the emotion determined by the emotion engine. If a negative emotion is included, a response procedure requiring special attention is suggested.

[1520] Notification and implementation of proposals

[1521] The server sends the generated proposal to a notification system, which notifies staff in the relevant departments. Notification methods include push notifications via smartphone applications and internal messaging systems. The users (staff) then check the notified proposal and implement specific measures. This enables quick and appropriate responses that reflect customer feedback in real time.

[1522] Adding specific examples

[1523] As a concrete example, consider the case where a customer in a store says, "I don't know where the product is." This voice data is collected by a smartphone app and converted into text data, "I don't know where the product is," using a voice recognition engine. The server then analyzes this text data using NLP technology and an emotion engine, extracting the keywords "product," "location," and "don't know," and determining that the emotion is negative. This data is classified into the "store support" category and assigned to the relevant department. Finally, the server generates a suggestion, "You need guidance to the product location and prompt customer service," and notifies the staff.

[1524] Prompt Sentence Examples

[1525] Voice data is collected and converted into text data. The converted text data is analyzed using NLP technology to extract keywords, analyze sentiment, and identify topics. Based on the results, appropriate suggestions are generated and notified to the relevant departments.

[1526] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1527] Step 1: Collecting audio data

[1528] The device (smartphone) collects the customer's voice data through the application using a microphone. The input is the customer's voice data, and the output is the collected voice file. This voice file is sent directly to the server.

[1529] Step 2: Convert audio data to text

[1530] The server receives the transmitted audio file and converts it into text data using a speech recognition engine. The input is the audio file, and the output is the converted text data. Specifically, the server analyzes the audio data using Google's speech recognition API and generates the corresponding text data.

[1531] Step 3: Save the text data

[1532] The server stores the converted text data in a database. The input of this process is the text data obtained in the previous step, and the output is the stored text data. A database management system is used to efficiently store the text data.

[1533] Step 4: Analyze and classify text data

[1534] The server analyzes the stored text data using natural language processing (NLP) technology and a sentiment analysis engine. Specifically, it uses the Hugging Face BERT model to extract keywords, analyze sentiment, and identify topics from the text data. The input is the text data, and the output is the extracted keywords, sentiment classification, and topic identification results.

[1535] Step 5: Allocate data

[1536] The server automatically assigns data to the appropriate department based on the analyzed and classified data. The input is keywords, sentiment classification, and topic identification results, and the output is department assignment information. This is done based on pre-defined categories.

[1537] Step 6: Generate proposals

[1538] The server generates appropriate proposals based on the assigned data. The input is the assignment information to the corresponding department, and the output is the generated proposal. The proposals are automatically created using an algorithm and include appropriate countermeasures.

[1539] Step 7: Notification of proposal

[1540] The server sends the generated proposal to a notification system, which notifies staff in the relevant departments. The input is the proposal, and the output is the notified information. Notification methods include push notifications on smartphones and internal messaging systems.

[1541] Step 8: Implementing customer interactions

[1542] The user (staff member) checks the notified proposal and takes specific action. The input is the notified proposal, and the output is the result of the action taken. This enables quick and appropriate action that reflects customer feedback in real time.

[1543] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1544] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1545] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1546] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1547] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1548] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1549] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1550] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1551] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1552] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1553] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1554] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1555] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1556] 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.

[1557] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1558] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1559] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1560] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1561] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1562] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1563] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1564] The following is further disclosed regarding the above embodiment.

[1565] (Claim 1)

[1566] a means for collecting customer voice data;

[1567] means for converting the voice data into text data;

[1568] means for analyzing and classifying said text data;

[1569] means for allocating the classified data to a corresponding department;

[1570] means for generating proposals based on the assigned data;

[1571] means for notifying the generated proposal content;

[1572] A system including:

[1573] (Claim 2)

[1574] 10. The system of claim 1, further comprising means for storing the collected audio data.

[1575] (Claim 3)

[1576] 2. The system according to claim 1, wherein the analysis means includes means for extracting keywords, analyzing sentiment, and identifying topics from text data using natural language processing techniques.

[1577] "Example 1"

[1578] (Claim 1)

[1579] a means for collecting customer voice data;

[1580] means for converting the voice data into text data;

[1581] A means for analyzing and classifying the text data by extracting keywords, analyzing sentiment, and identifying topics using natural language processing technology;

[1582] means for assigning the classified data to appropriate departments based on predefined categories;

[1583] A means for generating proposal content using a generative AI model based on the allocated data;

[1584] means for notifying the generated proposal content;

[1585] A system including:

[1586] (Claim 2)

[1587] 10. The system of claim 1, further comprising means for storing the collected audio data.

[1588] (Claim 3)

[1589] The system according to claim 1, further comprising means for inputting a prompt sentence to the generative AI model when generating the suggestion content.

[1590] "Application Example 1"

[1591] (Claim 1)

[1592] a means for collecting customer voice data;

[1593] means for converting the voice data into text data;

[1594] means for analyzing and classifying said text data;

[1595] means for allocating the classified data to a corresponding department;

[1596] means for generating proposals based on the assigned data;

[1597] means for notifying the generated proposal content;

[1598] means for extracting keywords from customer feedback using the voice data;

[1599] means for performing sentiment analysis using natural language processing techniques to classify categories based on the keywords;

[1600] A system including:

[1601] (Claim 2)

[1602] 10. The system of claim 1, further comprising means for storing the collected audio data.

[1603] (Claim 3)

[1604] 2. The system according to claim 1, wherein the analysis means includes means for extracting keywords, analyzing sentiment, and identifying topics from text data using natural language processing techniques.

[1605] "Example 2: Combining Emotion Engines"

[1606] (Claim 1)

[1607] a means for collecting customer voice data;

[1608] means for converting the voice data into text data;

[1609] means for analyzing and classifying said text data;

[1610] means for allocating the classified data to a corresponding department;

[1611] means for generating proposals based on the assigned data;

[1612] means for notifying the generated proposal content;

[1613] A means for the terminal to collect voice data in real time or in batch processing and transmit the data to a server;

[1614] A means for the server to convert voice data into text data using a voice recognition engine;

[1615] A means for the server to analyze the text data using natural language processing technology and an emotion engine;

[1616] A system including:

[1617] (Claim 2)

[1618] 10. The system of claim 1, further comprising means for storing the collected audio data.

[1619] (Claim 3)

[1620] 2. The system according to claim 1, wherein the analysis means includes means for extracting keywords, analyzing sentiment, and identifying topics from text data using natural language processing techniques.

[1621] "Application example 2 when combining emotion engines"

[1622] (Claim 1)

[1623] a means for collecting customer voice data;

[1624] means for converting the voice data into text data;

[1625] means for analyzing and classifying said text data;

[1626] means for allocating the classified data to a corresponding department;

[1627] means for generating proposals based on the assigned data;

[1628] means for notifying the generated proposal content;

[1629] a smartphone application means for responding to customers in a store in real time;

[1630] A system including:

[1631] (Claim 2)

[1632] 10. The system of claim 1, further comprising means for storing the collected audio data.

[1633] (Claim 3)

[1634] 2. The system according to claim 1, wherein the analysis means includes means for extracting keywords, analyzing sentiment, and identifying topics from text data using natural language processing techniques. [Explanation of symbols]

[1635] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for collecting customer voice data; means for converting the voice data into text data; means for analyzing and classifying said text data; means for allocating the classified data to a corresponding department; means for generating proposals based on the assigned data; means for notifying the generated proposal content; A system including:

2. The system of claim 1 further comprising means for storing the collected audio data.

3. The system according to claim 1 , wherein the analysis means includes means for extracting keywords, analyzing sentiment, and identifying topics from text data using natural language processing techniques.

Images