system

Abstract

We provide the system. [Solution] Means for obtaining audio data as a complaint, A means of converting acquired audio data into text data, A means of analyzing text data to classify and determine the severity of claims, A means for dynamically generating an appropriate response message based on a judgment, A means for providing the generated response message by speech synthesis, A means of transmitting commands when immediate response by a human operator is required, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] Claim handling in a call center has a problem that the mental and time burdens on operators are large. Handling with limited staff may result in variations in quality and may also lead to an increase in labor costs. Furthermore, in an environment where prompt and appropriate responses to particularly difficult claims or customer harassment are required, there is also a risk of operator turnover with conventional methods. The present invention aims to solve such problems and realize more efficient and high-quality claim handling.

Means for Solving the Problems

[0005] This invention provides a system that acquires audio data related to a specific claim and converts it into text data. The system analyzes the obtained text data to determine the type and severity of the claim. Furthermore, it automatically generates an appropriate response message based on this determination and provides it to the user. This system improves the quality of claim handling by monitoring the user's problem resolution and sending alerts to operators as needed. In addition, it has a function to determine and execute follow-up actions, reducing the burden on operators and enabling efficient operation.

[0006] "Voice data" refers to the audio information itself recorded during a call with a user, and is digital data acquired for the purpose of analyzing the content of the complaint.

[0007] "Text data" refers to text information created by converting audio data into written form using speech recognition technology, and is used to analyze the details of a claim.

[0008] "Analysis" is a series of processes that extract useful information from text data and make appropriate decisions based on that information.

[0009] "Types of complaints" refer to the content and categories of customer complaints and dissatisfaction, and are classified in order to determine the appropriate response.

[0010] "Severity" is a measure that indicates the scope and urgency of a complaint, and is a factor in determining the priority of responses.

[0011] A "response message" is a user-facing response generated based on system judgment, and includes explanations and solutions to customer concerns.

[0012] "Follow-up actions" are additional measures or follow-up tasks that should be taken after a complaint has been handled, and are necessary countermeasures to maintain customer satisfaction.

[0013] "Sending an alert to an operator" means that the system will notify an operator to take notice when a specific situation occurs. [Brief explanation of the drawing]

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

Embodiments for Carrying Out the Invention

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

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

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

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

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

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

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

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

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

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

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

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

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

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

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

[0035] This invention provides a system for streamlining customer complaint handling in call centers. This system combines speech recognition, natural language processing, and machine learning models, and operates in the following manner.

[0036] When a user calls a call center, the terminal sends the audio as digital audio data to the server. The server receives this audio data and converts it into text data using a speech recognition engine. The server then uses natural language processing techniques to analyze the text data in detail. The purpose of this analysis is to identify the types of complaints and emotional expressions contained in the text and to assess their severity based on these.

[0037] The analysis results are input into an AI model for handling complaints. The server utilizes the AI ​​model to generate the optimal response message based on similar past cases and statistical data. The response message is dynamically customized according to the type and severity of the complaint before being delivered to the user.

[0038] The terminal receives a response message sent from the server and conveys it to the user using speech synthesis technology. The content of the response varies widely, including confirmation of product description, apologies, and refund offers. If follow-up actions are to be taken, it may also include related details and schedules.

[0039] Furthermore, the server tracks the response results in real time, preparing for user reactions and any additional complaints. If the problem is not resolved, the server immediately sends an alert to the operator, requesting direct action.

[0040] As a concrete example, consider a case where a user complains about a delay in their order. After the terminal receives the audio, the server transcribes the audio into text and performs analysis to identify the cause of the complaint. The AI ​​model quickly identifies the keyword "delay" and, referencing effective responses in similar situations in the past, automatically generates a message explaining the reason for the delay and an apology. This ensures fast and consistent service, improving user satisfaction and reducing the burden on operators.

[0041] The following describes the processing flow.

[0042] Step 1:

[0043] A user makes a call to the call center, and the call begins. The device receives the call and records it as digital audio.

[0044] Step 2:

[0045] The terminal streams the recorded audio data to the server. The server stores the received audio data.

[0046] Step 3:

[0047] The server uses a speech recognition engine to convert the audio data into text data. This converts the audio content into a text format suitable for analysis.

[0048] Step 4:

[0049] The server applies natural language processing techniques to analyze the converted text data. The goal here is to identify the content of the complaint, the sentiment behind it, and important keywords.

[0050] Step 5:

[0051] Based on the analysis results, the server inputs the data into an AI model to determine the type and severity of the claim. This AI model compares the data with past data to recognize patterns.

[0052] Step 6:

[0053] The server generates the optimal response message for the user based on the AI ​​model's judgment. This message is dynamically assembled based on a predefined template.

[0054] Step 7:

[0055] The server sends the generated response message to the terminal. The terminal either reads this message aloud using speech synthesis technology or displays it as text.

[0056] Step 8:

[0057] The server monitors user responses, obtains follow-up actions as needed, and determines the execution plan.

[0058] Step 9:

[0059] If the server determines that a problem remains unresolved, or if an anomaly is detected in the system, it sends an alert to the operator. This allows the operator to intervene when necessary.

[0060] (Example 1)

[0061] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0062] Handling customer complaints in call centers is extremely diverse and requires both speed and accuracy, but traditional methods make it difficult to do so efficiently. Furthermore, it is necessary to reduce the burden on workers while simultaneously improving customer satisfaction. To address this challenge, there is a need for methods that automate speech conversion and analysis, as well as automate response generation.

[0063] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0064] In this invention, the server includes an information processing device for acquiring audio data, an acoustic analysis device for converting the acquired audio data into text data, and a language processing device for analyzing the text data and determining the type and importance of the information. This enables rapid and accurate analysis and automatic generation of responses to customer complaints.

[0065] An "information processing device for acquiring voice data" is a device that captures voice data from phone calls and other sources and converts it into a format usable within the system.

[0066] An "acoustic analysis device that converts acquired audio data into text data" is a device that uses speech recognition technology to perform the process of converting audio signals into text information.

[0067] A "language processing device that analyzes text data to determine the type and importance of information" is a device that utilizes natural language processing technology to extract and determine meaning, sentiment, and importance from text data.

[0068] A "machine learning device that generates appropriate response information based on judgment" is a device that creates the optimal response message according to the situation, based on past data and learning algorithms.

[0069] A "communication device that provides response information to a user" is a device that provides a means of transmitting generated response information to the user in voice or text format.

[0070] The "function to monitor the results of the response and send an alert to the worker if the problem is not resolved" means that the system tracks the processing results and user responses, and directly notifies a human if the problem remains unresolved under pre-set conditions.

[0071] This invention is a system primarily aimed at automating and streamlining customer complaint handling in call centers. The system processes voice data, analyzes text data using natural language processing technology, and provides the user with the most appropriate response. A detailed configuration is shown below.

[0072] The server acquires the voice spoken by the user through the terminal. The terminal converts the voice data into a digital format and transmits it to the server. The server uses speech recognition software (e.g., a general speech recognition engine) to convert this voice data into text data. The transcription of the voice provides the basic information necessary to accurately determine the type and severity of the claim.

[0073] This analysis uses specific natural language processing libraries (for example, open-source natural language processing libraries). In the sentiment analysis stage, sentiment keywords are extracted from the text to understand the user's emotional state. Furthermore, the server uses machine learning models to generate response messages appropriate for similar cases based on the data collected so far.

[0074] Generative AI models are used to analyze historical data and produce the most effective responses. For example, in response to a customer complaint about a delayed product, the AI ​​model appropriately constructs a message that includes an apology for the delay and details of the reason for the delay. This response is customized to the content of the complaint before being delivered to the user.

[0075] The terminal can use text-to-speech technology (e.g., a common speech synthesis system) to provide users with feedback in a natural voice based on text messages provided by the server. This process allows users to receive quicker complaint resolution, leading to increased satisfaction.

[0076] An example of a prompt might be, "Please provide information regarding the product delay and generate an apology voice message." Using this prompt, the system can demonstrate its ability to generate an appropriate response to a specific complaint.

[0077] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0078] Step 1:

[0079] When a user makes a phone call, the device receives the call and generates audio data in real time. The input is the user's voice, and the output is digital audio data. The specific action performed by the device is to collect the voice via the microphone and record it in digital format.

[0080] Step 2:

[0081] The terminal sends the generated audio data to the server. The input is the digital audio data acquired in the previous step, and the output is a compressed audio file sent to the server. The specific actions taken to reach the server involve the terminal compressing the data and sending it using a secure communication protocol.

[0082] Step 3:

[0083] The server uses speech recognition software to convert speech data into text data. The input is digital speech data received from the terminal, and the output is text data that transcribes the speech into written form. Specifically, the server applies a noise reduction filter and uses a language model to perform the process of converting speech to text.

[0084] Step 4:

[0085] The server analyzes the obtained text data using natural language processing techniques to determine the type and sentiment of the claim. The input is text data, and the output is an evaluation of the claim's attributes and sentiment. Specifically, the server performs grammatical structure analysis and key phrase extraction to identify important information within the text.

[0086] Step 5:

[0087] The server uses a generative AI model to generate the optimal response message based on the analysis results. The input is the analyzed claims attributes and emotions, and the output is the response text to be conveyed to the user. The server's specific operation is for the model to select a response template based on past data and incorporate it into specific wording.

[0088] Step 6:

[0089] The generated response message is sent to the terminal, which then uses speech synthesis technology to deliver the response message to the user verbally. The input is the response text from the server, and the output is the voice response delivered to the user. Specifically, the terminal uses an appropriate speech synthesis engine to play the message in a natural-sounding voice.

[0090] Step 7:

[0091] The server monitors user responses and evaluates the outcome of the response. Inputs are additional voice responses and evaluation feedback from the user, and outputs are decisions on the next action based on the results. Specifically, in this step, the server performs voice analysis in real time and activates an alarm system if the problem remains unresolved.

[0092] (Application Example 1)

[0093] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0094] In autonomously moving machines, it is essential to recognize user complaints and operational dissatisfaction in real time and respond quickly and accurately. Conventional systems have sometimes resulted in delays in complaint handling, degrading the user experience. Furthermore, it has been difficult to immediately propose improvements to the route and operation, hindering efficient operation. This invention solves these problems.

[0095] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0096] In this invention, the server includes means for acquiring voice data as a claim, means for converting the acquired voice data into text data, means for analyzing the text data to determine the classification and severity of the claim, means for dynamically generating an appropriate response message based on the determination, means for providing the generated response message by speech synthesis, and means for transmitting a command when immediate response by a human operator is required. This enables real-time handling of claims and suggestions for improving routes and operations.

[0097] "Audio data" refers to data that represents information generated as sound in a digital format, and is subject to analysis and conversion.

[0098] "Text data" refers to data obtained by converting audio data into string information, and is used as a foundation for natural language processing.

[0099] "Claim classification" is the process of identifying claims that belong to a specific type or category by analyzing text data.

[0100] "Severity" is a measure used to assess the magnitude and urgency of the impact a complaint has on business operations or services.

[0101] A "response message" is a message provided to the user that shows an appropriate reaction to a complaint.

[0102] "Speech synthesis" is a technology that outputs generated text-based response messages as speech that closely resembles a human voice.

[0103] A "human operator" is a human worker who intervenes and takes appropriate action when autonomous machine operation is difficult.

[0104] "Sending a command" refers to the act of notifying a human operator to take action when specific conditions occur.

[0105] This invention provides a system for autonomously moving machines to receive and immediately respond to user complaints during operation. An embodiment of this system is described below.

[0106] The server acquires audio data transmitted from vehicles and mobile devices. A speech recognition engine is used to convert this audio data into text data. Specifically, Google® Speech-to-Text and Amazon Transcribe can be used as speech recognition APIs. The converted text data is analyzed using natural language processing libraries (e.g., spaCy and NLTK) to determine the classification and severity of the claims.

[0107] Next, a generative AI model running on a cloud server dynamically generates the optimal response message based on the analysis results. This AI model learns from similar past claim data and presents an action plan in real time to resolve the user's problem. The generated response message is delivered to the user via voice from inside the vehicle or on a device using speech synthesis technology (e.g., Google Text-to-Speech).

[0108] Furthermore, the server can suggest improvements to the operating path and maneuvers of the autonomously moving machine. For example, if a user complains that "the brakes are too abrupt," the server can refer to past data and provide real-time feedback on improved braking operation.

[0109] Furthermore, if the problem is not resolved, it has the function to generate a warning and send a command to a human operator. This enables a quick response and improves the user experience.

[0110] Prompt example:

[0111] "What information is needed to analyze user comments and generate optimal feedback?"

[0112] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0113] Step 1:

[0114] The terminal acquires voice data from the user. The acquired voice data is converted into a digital format and sent to the server. This allows the voice information to be obtained in a digital format suitable for analysis.

[0115] Step 2:

[0116] The server converts the received audio data into text data using a speech recognition engine. For example, it uses Google Speech-to-Text to analyze the format of the audio waveform and convert it into text. This output text obtained from the input audio is then used for subsequent analysis.

[0117] Step 3:

[0118] The server analyzes text data using natural language processing techniques to classify and determine the severity of claims. It utilizes spaCy and NLTK to analyze the text content and extract specific keywords (e.g., sudden braking, delay). The analysis output based on the input text is the specific category of the claim and its evaluation.

[0119] Step 4:

[0120] Based on the analysis of the complaint, the server uses a generative AI model to generate the optimal response message. It refers to previously learned similar cases and statistical data to prepare a dynamic reply to the user. This process is text-based, and the AI ​​model generates an output message incorporating information obtained from the input analysis.

[0121] Step 5:

[0122] The server converts the generated response message into speech using a speech synthesis engine (e.g., Google Text-to-Speech) and provides it to the user via the device. By outputting text-based messages as speech and presenting them directly to the user, it enables easily understandable communication.

[0123] Step 6:

[0124] If the problem is not resolved, the server generates a warning and sends instructions to a human operator. If action is required, it generates an alert and sends a notification to the appropriate management system to enable immediate human intervention.

[0125] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0126] This invention provides a system for efficiently and effectively handling customer complaints in a call center. This system has an integrated configuration of speech recognition, natural language processing, an emotion engine, and a machine learning model.

[0127] When a user makes a call, the terminal receives the voice call and sends it to the server as digital audio data. The server converts the received audio data into text data using a speech recognition engine. The server then applies natural language processing techniques to this converted text data to analyze the type of claim, key keywords, and severity.

[0128] Furthermore, the server incorporates an emotion engine that can analyze user emotions from text data. The emotion engine analyzes the user's vocabulary and tone of voice to recognize emotions such as satisfaction, dissatisfaction, and anger in real time.

[0129] The analyzed data is processed by an AI model to dynamically generate response messages tailored to the user's complaint and emotional state. When these generated response messages are delivered to the user, they are adjusted according to their emotions, ensuring appropriate tone and content for effective communication.

[0130] For example, if a user expresses anger about a defective product, the device records the audio and sends it to a server. The server analyzes the audio, identifies the emotion of anger, and generates a message that includes an apology and a replacement offer. The device then uses speech synthesis to deliver this message to the user. The server also determines if follow-up is necessary and sends an alert to an operator to prompt a quick response.

[0131] The server continuously monitors the response results and improves and adjusts the response measures as needed. This system is expected to improve the quality of complaint handling and increase customer satisfaction.

[0132] The following describes the processing flow.

[0133] Step 1:

[0134] A user makes a call to a call center, and the device receives the voice call. The call content is transmitted to the server in real time as voice data.

[0135] Step 2:

[0136] The server activates the speech recognition engine and converts the received audio data into text data. This makes the audio content available in text format.

[0137] Step 3:

[0138] The server analyzes the text data using natural language processing. The analysis extracts the type of complaint and keywords, while the sentiment engine simultaneously identifies the user's emotions.

[0139] Step 4:

[0140] The emotion engine evaluates the user's emotional state (e.g., anger, dissatisfaction, satisfaction) in real time based on the tone and word choice of their speech. Data including an emotion score is then generated.

[0141] Step 5:

[0142] The server inputs the analysis results and emotional data into the AI ​​model. The AI ​​model generates the optimal response message based on the characteristics of the complaint and the user's emotional state.

[0143] Step 6:

[0144] The server sends the generated response message to the terminal. The terminal uses speech synthesis technology to provide the user with the response message in either voice or text format.

[0145] Step 7:

[0146] The server monitors user responses and determines whether follow-up action is needed. If necessary, it sends a notification to the operator.

[0147] Step 8:

[0148] If the user submits another complaint or feedback, the server will initiate a new analysis and repeat the process. This process will continue until the user's problem is resolved and they are satisfied.

[0149] (Example 2)

[0150] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0151] Handling complaints in a call center presents a challenging task, requiring efficient and effective responses that are sensitive to the user's emotions. It is also crucial to establish an appropriate tracking system to quickly resolve user dissatisfaction. Furthermore, accurately monitoring the outcome of responses and taking immediate action if the problem remains unresolved is essential.

[0152] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0153] In this invention, the server includes means for acquiring voice information, means for converting the acquired voice information into text information, means for analyzing the text information to determine the type and severity of the claim, means for adjusting the generated response message to match the user's emotions, means for providing the adjusted response message by speech synthesis, means for determining a tracking function for the user based on the type and severity of the claim and the user's emotions, and further means for monitoring the processing results and sending a notification to the person in charge if the problem is not resolved. This enables appropriate responses that are in line with the user's emotions and rapid problem resolution in claim handling.

[0154] "Voice information" refers to information, including the speaker's voice, obtained through telephones or communication systems.

[0155] "Textual information" refers to information in a readable string format, obtained by converting audio information using speech recognition technology.

[0156] "Complaint type" refers to a category used to classify the content of user dissatisfaction or requests, which is analyzed from the acquired textual information.

[0157] "Severity" refers to the degree of urgency or importance that requires action, as determined based on the type of claim.

[0158] A "response message" is the content of the reply provided to the user based on the results of analyzing voice and text information.

[0159] "Speech synthesis" is a technology that mechanically generates speech based on textual information and provides information to the user audibly.

[0160] The "tracking function" is a feature that monitors the progress of handling complaints and prompts further action as needed.

[0161] A "notification" is a means of communication used to warn the person in charge if the problem is not resolved based on the results of the response.

[0162] This invention provides a system for efficiently and effectively handling customer complaints in a call center. This system integrates speech recognition technology, natural language processing technology, sentiment analysis technology, and machine learning models.

[0163] When a user makes a phone call, the device receives the voice call, converts the voice data into a digital format, and sends it to the server. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice data into text data. The converted text data is then analyzed using natural language processing techniques (e.g., spaCy or NLTK) to identify the type of claim, key keywords, and severity.

[0164] The server incorporates an emotion engine (e.g., IBM Watson® Tone Analyzer) that analyzes user emotions from text data. By analyzing the user's vocabulary and tone of voice, emotions such as satisfaction, dissatisfaction, and anger are recognized in real time. The analyzed data is processed by an AI model (e.g., OpenAI® GPT-3®) to dynamically generate response messages tailored to the user's complaint and emotional state. These response messages are adjusted according to the emotion and delivered to the user in an appropriate tone and content.

[0165] For example, if a user expresses anger about a defective product, the device records the audio and sends it to a server. The server analyzes the audio, identifies the level of anger, and generates a message that includes an apology and a replacement offer. The device then delivers this message to the user using speech synthesis (e.g., Amazon Polly). The server can also determine if follow-up is necessary and send an alert to an operator to encourage a quick response.

[0166] An example of a prompt is, "Generate the optimal response based on the user's emotions and complaint details." This prompt prompts the AI ​​model to generate the most appropriate response message. This is expected to improve the quality of complaint handling and increase customer satisfaction.

[0167] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0168] Step 1:

[0169] The user makes a phone call. The terminal receives the voice call and converts the voice data into a digital format. The input is the user's voice, and the output is digital voice data. The terminal sends this digital voice data to the server.

[0170] Step 2:

[0171] The server inputs the received digital audio data into the speech recognition engine and converts it into text data. Here, the input is digital audio data, and the output is text data. The server then verifies whether the conversion was successful.

[0172] Step 3:

[0173] The server inputs the converted text data into a natural language processing engine. This engine analyzes the text data to identify the type of claim, key keywords, and severity. The input is text data, and the output is the analyzed category, keyword, and severity information.

[0174] Step 4:

[0175] The server inputs text data into its emotion analysis engine to analyze the user's emotions. The input is the text data obtained in step 3, and the output is information about emotions. Specifically, emotions such as satisfaction, dissatisfaction, and anger are identified.

[0176] Step 5:

[0177] The server inputs the analysis results into a generation AI model and generates a response message tailored to the user's complaint and emotional state. The input includes information on category, keywords, severity, and emotion, while the output is the response message.

[0178] Step 6:

[0179] The server adjusts the response message to match the user's emotions. In this step, the generated message and emotional information are used as input to output an optimized response message. Specifically, the tone and content of the message are adjusted.

[0180] Step 7:

[0181] The terminal inputs a pre-arranged response message into a speech synthesis engine, converts it into speech format, and delivers it to the user. The input is an optimized text message, and the output is synthesized speech. The terminal then plays this speech message to the user.

[0182] Step 8:

[0183] The server continuously monitors the response results. If the problem is not resolved, it sends a notification to the responsible person as needed to prompt action. In this step, it receives the processing results as input and outputs an alert if action is required.

[0184] (Application Example 2)

[0185] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0186] In voice-based request processing, it is essential to accurately assess the user's emotions and urgency, and to respond quickly and appropriately. However, conventional systems often fail to adequately analyze emotions or assess urgency, leading to decreased customer satisfaction. This is especially true in areas where security is paramount, requiring rapid risk prediction and appropriate response. Therefore, there is a need for technology that can highly analyze voice information and enable appropriate responses and follow-up.

[0187] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0188] In this invention, the server includes a device for acquiring voice information, a device for converting the acquired voice information into text information, a device for analyzing the text information to determine the type and severity of the request, and a device for analyzing emotions to determine the urgency. This enables a quick and appropriate response based on the user's emotions and urgency.

[0189] "Audio information" refers to data based on sound waves acquired from an input device.

[0190] "Textual information" refers to string data that is obtained by converting audio information through a recognition device.

[0191] A "requirement" is a concept that describes the problems or needs that a user expresses to a system.

[0192] "Severity" is a measure that indicates the level of importance and urgency of a request.

[0193] "Emotions" are elements that express the user's subjective mental state.

[0194] "Urgency" is a measure that indicates whether an immediate response is necessary to a particular situation.

[0195] A "device" is a collection of hardware and software used to process, analyze, and interpret audio and text information.

[0196] The system implementing this invention is designed to process requests using voice. When a user inputs voice information, the system's terminal collects that voice information and transmits it to a server. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice information into text information. This text information is then analyzed by a natural language processing engine (e.g., Google Natural Language API).

[0197] Based on the analyzed information, the server determines the type and severity of the request. In addition, a sentiment analysis API (e.g., IBM Watson Tone Analyzer) is used to identify the user's emotions and their urgency. This data is input into a generative AI model configured for automated response generation, which then generates appropriate response information. The generated response information is then converted back into speech using a speech synthesis engine (e.g., Amazon Polly) and provided to the user.

[0198] This system allows users to, for example, voice-input a statement like, "I've recently been feeling unsafe," and the system analyzes their feelings of anxiety. If it determines that a security patrol is necessary, measures are immediately taken. By taking appropriate follow-up actions according to the user's situation, improved user safety can be expected.

[0199] A concrete example of a prompt sentence corresponding to the generative AI model is, "Generate the most appropriate response based on the user's voice." This invention contributes to providing a quick response and a sense of security.

[0200] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0201] Step 1:

[0202] The device acquires voice information from the user. Voice information is collected when the user speaks into the smartphone, and this voice is stored digitally by the device. This voice information becomes the initial input data.

[0203] Step 2:

[0204] The device sends the acquired audio information to the server. The audio data is uploaded to the server via the internet. This transfer to the server enables the start of the next process.

[0205] Step 3:

[0206] The server uses a speech recognition engine to convert speech information into text. Specifically, it utilizes the Google Cloud Speech-to-Text API to convert speech data into text data. This process records the information obtained from the speech as text.

[0207] Step 4:

[0208] The server analyzes the text information using a natural language processing engine. Through the Google Natural Language API, it identifies the type and severity of the request from the text information. It then extracts important keywords from the input text data and performs a process to understand its context.

[0209] Step 5:

[0210] The server uses a sentiment analysis API to analyze the sentiment of text information. IBM Watson Tone Analyzer is used to determine the user's emotional state. Emotions are classified based on the phrasing and tone of the text data.

[0211] Step 6:

[0212] The server generates appropriate response information using a generated AI model. Based on the type, severity, and emotional state of the request obtained in the previous step, the AI ​​generates the optimal message using data including prompt text as input.

[0213] Step 7:

[0214] The server converts the generated response information into speech using a speech synthesis engine. Amazon Polly is then used to convert the generated text response back into speech. The result is presented to the user as a voice response.

[0215] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0216] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0217] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0218] [Second Embodiment]

[0219] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0220] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0221] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0222] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0223] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0224] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0225] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0226] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0227] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0228] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0229] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0230] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0231] This invention provides a system for streamlining customer complaint handling in call centers. This system combines speech recognition, natural language processing, and machine learning models, and operates in the following manner.

[0232] When a user calls a call center, the terminal sends the audio as digital audio data to the server. The server receives this audio data and converts it into text data using a speech recognition engine. The server then uses natural language processing techniques to analyze the text data in detail. The purpose of this analysis is to identify the types of complaints and emotional expressions contained in the text and to assess their severity based on these.

[0233] The analysis results are input into an AI model for handling complaints. The server utilizes the AI ​​model to generate the optimal response message based on similar past cases and statistical data. The response message is dynamically customized according to the type and severity of the complaint before being delivered to the user.

[0234] The terminal receives a response message sent from the server and conveys it to the user using speech synthesis technology. The content of the response varies widely, including confirmation of product description, apologies, and refund offers. If follow-up actions are to be taken, it may also include related details and schedules.

[0235] Furthermore, the server tracks the response results in real time, preparing for user reactions and any additional complaints. If the problem is not resolved, the server immediately sends an alert to the operator, requesting direct action.

[0236] As a concrete example, consider a case where a user complains about a delay in their order. After the terminal receives the audio, the server transcribes the audio into text and performs analysis to identify the cause of the complaint. The AI ​​model quickly identifies the keyword "delay" and, referencing effective responses in similar situations in the past, automatically generates a message explaining the reason for the delay and an apology. This ensures fast and consistent service, improving user satisfaction and reducing the burden on operators.

[0237] The following describes the processing flow.

[0238] Step 1:

[0239] A user makes a call to the call center, and the call begins. The device receives the call and records it as digital audio.

[0240] Step 2:

[0241] The terminal streams the recorded audio data to the server. The server stores the received audio data.

[0242] Step 3:

[0243] The server uses a speech recognition engine to convert the audio data into text data. This converts the audio content into a text format suitable for analysis.

[0244] Step 4:

[0245] The server applies natural language processing techniques to analyze the converted text data. The goal here is to identify the content of the complaint, the sentiment behind it, and important keywords.

[0246] Step 5:

[0247] Based on the analysis results, the server inputs the data into an AI model to determine the type and severity of the claim. This AI model compares the data with past data to recognize patterns.

[0248] Step 6:

[0249] The server generates the optimal response message for the user based on the AI ​​model's judgment. This message is dynamically assembled based on a predefined template.

[0250] Step 7:

[0251] The server sends the generated response message to the terminal. The terminal either reads this message aloud using speech synthesis technology or displays it as text.

[0252] Step 8:

[0253] The server monitors user responses, obtains follow-up actions as needed, and determines the execution plan.

[0254] Step 9:

[0255] If the server determines that a problem remains unresolved, or if an anomaly is detected in the system, it sends an alert to the operator. This allows the operator to intervene when necessary.

[0256] (Example 1)

[0257] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0258] Handling customer complaints in call centers is extremely diverse and requires both speed and accuracy, but traditional methods make it difficult to do so efficiently. Furthermore, it is necessary to reduce the burden on workers while simultaneously improving customer satisfaction. To address this challenge, there is a need for methods that automate speech conversion and analysis, as well as automate response generation.

[0259] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0260] In this invention, the server includes an information processing device for acquiring audio data, an acoustic analysis device for converting the acquired audio data into text data, and a language processing device for analyzing the text data and determining the type and importance of the information. This enables rapid and accurate analysis and automatic generation of responses to customer complaints.

[0261] An "information processing device for acquiring voice data" is a device that captures voice data from phone calls and other sources and converts it into a format usable within the system.

[0262] An "acoustic analysis device that converts acquired audio data into text data" is a device that uses speech recognition technology to perform the process of converting audio signals into text information.

[0263] A "language processing device that analyzes text data to determine the type and importance of information" is a device that utilizes natural language processing technology to extract and determine meaning, sentiment, and importance from text data.

[0264] A "machine learning device that generates appropriate response information based on judgment" is a device that creates the optimal response message according to the situation, based on past data and learning algorithms.

[0265] A "communication device that provides response information to a user" is a device that provides a means of transmitting generated response information to the user in voice or text format.

[0266] The "function to monitor the results of the response and send an alert to the worker if the problem is not resolved" means that the system tracks the processing results and user responses, and directly notifies a human if the problem remains unresolved under pre-set conditions.

[0267] This invention is a system primarily aimed at automating and streamlining customer complaint handling in call centers. The system processes voice data, analyzes text data using natural language processing technology, and provides the user with the most appropriate response. A detailed configuration is shown below.

[0268] The server acquires the voice spoken by the user through the terminal. The terminal converts the voice data into a digital format and transmits it to the server. The server uses speech recognition software (e.g., a general speech recognition engine) to convert this voice data into text data. The transcription of the voice provides the basic information necessary to accurately determine the type and severity of the claim.

[0269] This analysis uses specific natural language processing libraries (for example, open-source natural language processing libraries). In the sentiment analysis stage, sentiment keywords are extracted from the text to understand the user's emotional state. Furthermore, the server uses machine learning models to generate response messages appropriate for similar cases based on the data collected so far.

[0270] Generative AI models are used to analyze historical data and produce the most effective responses. For example, in response to a customer complaint about a delayed product, the AI ​​model appropriately constructs a message that includes an apology for the delay and details of the reason for the delay. This response is customized to the content of the complaint before being delivered to the user.

[0271] The terminal can use text-to-speech technology (e.g., a common speech synthesis system) to provide users with feedback in a natural voice based on text messages provided by the server. This process allows users to receive quicker complaint resolution, leading to increased satisfaction.

[0272] An example of a prompt might be, "Please provide information regarding the product delay and generate an apology voice message." Using this prompt, the system can demonstrate its ability to generate an appropriate response to a specific complaint.

[0273] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0274] Step 1:

[0275] When a user makes a phone call, the device receives the call and generates audio data in real time. The input is the user's voice, and the output is digital audio data. The specific action performed by the device is to collect the voice via the microphone and record it in digital format.

[0276] Step 2:

[0277] The terminal sends the generated voice data to the server. The input is the digital voice data obtained in the previous step, and the output is the compressed voice file sent to the server. As specific operations until reaching the server, the terminal performs data compression and sends the data using a secure communication protocol.

[0278] Step 3:

[0279] The server converts the voice data into text data using voice recognition software. The input is the digital voice data received from the terminal, and the output is the text data obtained by transcribing the voice into characters. As specific operations, the server applies a noise removal filter and executes a process of converting the voice into text using a language model.

[0280] Step 4:

[0281] The server analyzes the obtained text data using natural language processing technology and determines the type and sentiment of the claim. The input is the text data, and the output is the evaluation result of the claim's attributes and sentiment. As specific operations, the server performs syntactic analysis of the text and extracts key phrases to identify important information in the text.

[0282] Step 5:

[0283] Based on the analysis results, the server generates an optimal response message using a generative AI model. The input is the analyzed claim attributes and sentiment, and the output is the response text to be conveyed to the user. The specific operation of the server is that the model selects a response template based on past data and incorporates it into specific language.

[0284] Step 6:

[0285] The generated response message is sent to the terminal, and the terminal uses voice synthesis technology to convey the response message to the user in voice. The input is the response text from the server, and the output is the voice response conveyed to the user. The specific operation is that the terminal uses an appropriate voice synthesis engine to play the message in natural voice.

[0286] Step 7:

[0287] The server monitors the user's reaction and evaluates the result of the response. The input is the user's additional voice response or evaluation feedback, and the output is the determination of the next action based on the result. The specific operation in this step is that the server performs voice analysis in real time and activates the alarm system if the problem remains unsolved.

[0288] (Application Example 1)

[0289] Next, Application Example 1 will be described. In the following description, the data processing device 12 is referred to as the "server", and the smart glasses 214 are referred to as the "terminal".

[0290] In a machine during autonomous movement, it is required to recognize claims from users and dissatisfaction during operation in real time and respond quickly and accurately. In conventional systems, claim handling was delayed, which could reduce the user experience. Also, it was difficult to immediately present improvement plans for the operation route and operations, hindering efficient operation. The present invention solves these problems.

[0291] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0292] In this invention, the server includes means for acquiring voice data as a claim, means for converting the acquired voice data into text data, means for analyzing the text data to determine the classification and severity of the claim, means for dynamically generating an appropriate response message based on the determination, means for providing the generated response message by speech synthesis, and means for transmitting a command when immediate response by a human operator is required. This enables real-time handling of claims and suggestions for improving routes and operations.

[0293] "Audio data" refers to data that represents information generated as sound in a digital format, and is subject to analysis and conversion.

[0294] "Text data" refers to data obtained by converting audio data into string information, and is used as a foundation for natural language processing.

[0295] "Claim classification" is the process of identifying claims that belong to a specific type or category by analyzing text data.

[0296] "Severity" is a measure used to assess the magnitude and urgency of the impact a complaint has on business operations or services.

[0297] A "response message" is a message provided to the user that shows an appropriate reaction to a complaint.

[0298] "Speech synthesis" is a technology that outputs generated text-based response messages as speech that closely resembles a human voice.

[0299] A "human operator" is a human worker who intervenes and takes appropriate action when autonomous machine operation is difficult.

[0300] "Sending a command" refers to the act of notifying a human operator to take action when specific conditions occur.

[0301] This invention provides a system for autonomously moving machines to receive and immediately respond to user complaints during operation. An embodiment of this system is described below.

[0302] The server acquires audio data transmitted from vehicles and mobile devices. A speech recognition engine is used to convert this audio data into text data. Specifically, Google Speech-to-Text and Amazon Transcribe can be used as speech recognition APIs. The converted text data is analyzed using natural language processing libraries (e.g., spaCy and NLTK) to classify the claims and determine their severity.

[0303] Next, a generative AI model running on a cloud server dynamically generates the optimal response message based on the analysis results. This AI model learns from similar past claim data and presents an action plan in real time to resolve the user's problem. The generated response message is delivered to the user via voice from inside the vehicle or on a device using speech synthesis technology (e.g., Google Text-to-Speech).

[0304] Furthermore, the server can suggest improvements to the operating path and maneuvers of the autonomously moving machine. For example, if a user complains that "the brakes are too abrupt," the server can refer to past data and provide real-time feedback on improved braking operation.

[0305] Furthermore, if the problem is not resolved, it has the function to generate a warning and send a command to a human operator. This enables a quick response and improves the user experience.

[0306] Prompt example:

[0307] "What information is needed to analyze user comments and generate optimal feedback?"

[0308] The flow of the specific process in Application Example 1 will be described using FIG. 12.

[0309] Step 1:

[0310] The terminal acquires voice data from the user. The acquired voice data is converted into a digital format and transmitted to the server. As a result, the voice information can be obtained in a digital format suitable for analysis.

[0311] Step 2:

[0312] The server converts the received voice data into text data using a voice recognition engine. For example, Google Speech-to-Text is used to analyze the format of the voice waveform and convert it into text format. The output text obtained from this input voice is used for subsequent analysis.

[0313] Step 3:

[0314] The server analyzes the text data using natural language processing techniques and determines the classification and severity of the claim. Utilizing spaCy or NLTK, the content of the text is analyzed to extract specific keywords (e.g., sudden brake, delay, etc.). The analysis output based on the input text is the specific category of the claim and its evaluation.

[0315] Step 4:

[0316] Based on the analysis result of the claim, the server generates an optimal response message using a generative AI model. Referring to previously learned similar cases and statistical data, a dynamic reply to the user is prepared. This process is performed textually, and the AI model generates an output message incorporating the information obtained from the input analysis.

[0317] Step 5:

[0318] The server converts the generated response message into speech using a speech synthesis engine (e.g., Google Text-to-Speech) and provides it to the user via the device. By outputting text-based messages as speech and presenting them directly to the user, it enables easily understandable communication.

[0319] Step 6:

[0320] If the problem is not resolved, the server generates a warning and sends instructions to a human operator. If action is required, it generates an alert and sends a notification to the appropriate management system to enable immediate human intervention.

[0321] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0322] This invention provides a system for efficiently and effectively handling customer complaints in a call center. This system has an integrated configuration of speech recognition, natural language processing, an emotion engine, and a machine learning model.

[0323] When a user makes a call, the terminal receives the voice call and sends it to the server as digital audio data. The server converts the received audio data into text data using a speech recognition engine. The server then applies natural language processing techniques to this converted text data to analyze the type of claim, key keywords, and severity.

[0324] Furthermore, the server incorporates an emotion engine that can analyze user emotions from text data. The emotion engine analyzes the user's vocabulary and tone of voice to recognize emotions such as satisfaction, dissatisfaction, and anger in real time.

[0325] The analyzed data is processed by an AI model to dynamically generate response messages tailored to the user's complaint and emotional state. When these generated response messages are delivered to the user, they are adjusted according to their emotions, ensuring appropriate tone and content for effective communication.

[0326] For example, if a user expresses anger about a defective product, the device records the audio and sends it to a server. The server analyzes the audio, identifies the emotion of anger, and generates a message that includes an apology and a replacement offer. The device then uses speech synthesis to deliver this message to the user. The server also determines if follow-up is necessary and sends an alert to an operator to prompt a quick response.

[0327] The server continuously monitors the response results and improves and adjusts the response measures as needed. This system is expected to improve the quality of complaint handling and increase customer satisfaction.

[0328] The following describes the processing flow.

[0329] Step 1:

[0330] A user makes a call to a call center, and the device receives the voice call. The call content is transmitted to the server in real time as voice data.

[0331] Step 2:

[0332] The server activates the speech recognition engine and converts the received audio data into text data. This makes the audio content available in text format.

[0333] Step 3:

[0334] The server analyzes the text data using natural language processing. The analysis extracts the type of complaint and keywords, while the sentiment engine simultaneously identifies the user's emotions.

[0335] Step 4:

[0336] The emotion engine evaluates the user's emotional state (e.g., anger, dissatisfaction, satisfaction) in real time based on the tone and word choice of their speech. Data including an emotion score is then generated.

[0337] Step 5:

[0338] The server inputs the analysis results and emotional data into the AI ​​model. The AI ​​model generates the optimal response message based on the characteristics of the complaint and the user's emotional state.

[0339] Step 6:

[0340] The server sends the generated response message to the terminal. The terminal uses speech synthesis technology to provide the user with the response message in either voice or text format.

[0341] Step 7:

[0342] The server monitors user responses and determines whether follow-up action is needed. If necessary, it sends a notification to the operator.

[0343] Step 8:

[0344] If the user submits another complaint or feedback, the server will initiate a new analysis and repeat the process. This process will continue until the user's problem is resolved and they are satisfied.

[0345] (Example 2)

[0346] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0347] Handling complaints in a call center presents a challenging task, requiring efficient and effective responses that are sensitive to the user's emotions. It is also crucial to establish an appropriate tracking system to quickly resolve user dissatisfaction. Furthermore, accurately monitoring the outcome of responses and taking immediate action if the problem remains unresolved is essential.

[0348] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0349] In this invention, the server includes means for acquiring voice information, means for converting the acquired voice information into text information, means for analyzing the text information to determine the type and severity of the claim, means for adjusting the generated response message to match the user's emotions, means for providing the adjusted response message by speech synthesis, means for determining a tracking function for the user based on the type and severity of the claim and the user's emotions, and further means for monitoring the processing results and sending a notification to the person in charge if the problem is not resolved. This enables appropriate responses that are in line with the user's emotions and rapid problem resolution in claim handling.

[0350] "Voice information" refers to information, including the speaker's voice, obtained through telephones or communication systems.

[0351] "Textual information" refers to information in a readable string format, obtained by converting audio information using speech recognition technology.

[0352] "Complaint type" refers to a category used to classify the content of user dissatisfaction or requests, which is analyzed from the acquired textual information.

[0353] "Severity" refers to the degree of urgency or importance that requires action, as determined based on the type of claim.

[0354] A "response message" is the content of the reply provided to the user based on the results of analyzing voice and text information.

[0355] "Speech synthesis" is a technology that mechanically generates speech based on textual information and provides information to the user audibly.

[0356] The "tracking function" is a feature that monitors the progress of handling complaints and prompts further action as needed.

[0357] A "notification" is a means of communication used to warn the person in charge if the problem is not resolved based on the results of the response.

[0358] This invention provides a system for efficiently and effectively handling customer complaints in a call center. This system integrates speech recognition technology, natural language processing technology, sentiment analysis technology, and machine learning models.

[0359] When a user makes a phone call, the device receives the voice call, converts the voice data into a digital format, and sends it to the server. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice data into text data. The converted text data is then analyzed using natural language processing techniques (e.g., spaCy or NLTK) to identify the type of claim, key keywords, and severity.

[0360] The server incorporates an emotion engine (e.g., IBM Watson Tone Analyzer) that analyzes user emotions from text data. By analyzing the user's vocabulary and tone of voice, emotions such as satisfaction, dissatisfaction, and anger are recognized in real time. The analyzed data is processed by an AI model (e.g., OpenAI GPT-3) to dynamically generate response messages tailored to the user's complaint and emotional state. These response messages are adjusted according to the emotion and delivered to the user in an appropriate tone and content.

[0361] For example, if a user expresses anger about a defective product, the device records the audio and sends it to a server. The server analyzes the audio, identifies the level of anger, and generates a message that includes an apology and a replacement offer. The device then delivers this message to the user using speech synthesis (e.g., Amazon Polly). The server can also determine if follow-up is necessary and send an alert to an operator to encourage a quick response.

[0362] An example of a prompt is, "Generate the optimal response based on the user's emotions and complaint details." This prompt prompts the AI ​​model to generate the most appropriate response message. This is expected to improve the quality of complaint handling and increase customer satisfaction.

[0363] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0364] Step 1:

[0365] The user makes a phone call. The terminal receives the voice call and converts the voice data into a digital format. The input is the user's voice, and the output is digital voice data. The terminal sends this digital voice data to the server.

[0366] Step 2:

[0367] The server inputs the received digital audio data into the speech recognition engine and converts it into text data. Here, the input is digital audio data, and the output is text data. The server then verifies whether the conversion was successful.

[0368] Step 3:

[0369] The server inputs the converted text data into a natural language processing engine. This engine analyzes the text data to identify the type of claim, key keywords, and severity. The input is text data, and the output is the analyzed category, keyword, and severity information.

[0370] Step 4:

[0371] The server inputs text data into its emotion analysis engine to analyze the user's emotions. The input is the text data obtained in step 3, and the output is information about emotions. Specifically, emotions such as satisfaction, dissatisfaction, and anger are identified.

[0372] Step 5:

[0373] The server inputs the analysis results into a generation AI model and generates a response message tailored to the user's complaint and emotional state. The input includes information on category, keywords, severity, and emotion, while the output is the response message.

[0374] Step 6:

[0375] The server adjusts the response message to match the user's emotions. In this step, the generated message and emotional information are used as input to output an optimized response message. Specifically, the tone and content of the message are adjusted.

[0376] Step 7:

[0377] The terminal inputs a pre-arranged response message into a speech synthesis engine, converts it into speech format, and delivers it to the user. The input is an optimized text message, and the output is synthesized speech. The terminal then plays this speech message to the user.

[0378] Step 8:

[0379] The server continuously monitors the response results. If the problem is not resolved, it sends a notification to the responsible person as needed to prompt action. In this step, it receives the processing results as input and outputs an alert if action is required.

[0380] (Application Example 2)

[0381] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0382] In voice-based request processing, it is essential to accurately assess the user's emotions and urgency, and to respond quickly and appropriately. However, conventional systems often fail to adequately analyze emotions or assess urgency, leading to decreased customer satisfaction. This is especially true in areas where security is paramount, requiring rapid risk prediction and appropriate response. Therefore, there is a need for technology that can highly analyze voice information and enable appropriate responses and follow-up.

[0383] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0384] In this invention, the server includes a device for acquiring voice information, a device for converting the acquired voice information into text information, a device for analyzing the text information to determine the type and severity of the request, and a device for analyzing emotions to determine the urgency. This enables a quick and appropriate response based on the user's emotions and urgency.

[0385] "Audio information" refers to data based on sound waves acquired from an input device.

[0386] "Textual information" refers to string data that is obtained by converting audio information through a recognition device.

[0387] A "requirement" is a concept that describes the problems or needs that a user expresses to a system.

[0388] "Severity" is a measure that indicates the level of importance and urgency of a request.

[0389] "Emotions" are elements that express the user's subjective mental state.

[0390] "Urgency" is a measure that indicates whether an immediate response is necessary to a particular situation.

[0391] A "device" is a collection of hardware and software used to process, analyze, and interpret audio and text information.

[0392] The system implementing this invention is designed to process requests using voice. When a user inputs voice information, the system's terminal collects that voice information and transmits it to a server. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice information into text information. This text information is then analyzed by a natural language processing engine (e.g., Google Natural Language API).

[0393] Based on the analyzed information, the server determines the type and severity of the request. In addition, a sentiment analysis API (e.g., IBM Watson Tone Analyzer) is used to identify the user's emotions and their urgency. This data is input into a generative AI model configured for automated response generation, which then generates appropriate response information. The generated response information is then converted back into speech using a speech synthesis engine (e.g., Amazon Polly) and provided to the user.

[0394] This system allows users to, for example, voice-input a statement like, "I've recently been feeling unsafe," and the system analyzes their feelings of anxiety. If it determines that a security patrol is necessary, measures are immediately taken. By taking appropriate follow-up actions according to the user's situation, improved user safety can be expected.

[0395] A concrete example of a prompt sentence corresponding to the generative AI model is, "Generate the most appropriate response based on the user's voice." This invention contributes to providing a quick response and a sense of security.

[0396] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0397] Step 1:

[0398] The device acquires voice information from the user. Voice information is collected when the user speaks into the smartphone, and this voice is stored digitally by the device. This voice information becomes the initial input data.

[0399] Step 2:

[0400] The device sends the acquired audio information to the server. The audio data is uploaded to the server via the internet. This transfer to the server enables the start of the next process.

[0401] Step 3:

[0402] The server uses a speech recognition engine to convert speech information into text. Specifically, it utilizes the Google Cloud Speech-to-Text API to convert speech data into text data. This process records the information obtained from the speech as text.

[0403] Step 4:

[0404] The server analyzes the text information using a natural language processing engine. Through the Google Natural Language API, it identifies the type and severity of the request from the text information. It then extracts important keywords from the input text data and performs a process to understand its context.

[0405] Step 5:

[0406] The server uses a sentiment analysis API to analyze the sentiment of text information. IBM Watson Tone Analyzer is used to determine the user's emotional state. Emotions are classified based on the phrasing and tone of the text data.

[0407] Step 6:

[0408] The server generates appropriate response information using a generated AI model. Based on the type, severity, and emotional state of the request obtained in the previous step, the AI ​​generates the optimal message using data including prompt text as input.

[0409] Step 7:

[0410] The server converts the generated response information into speech using a speech synthesis engine. Amazon Polly is then used to convert the generated text response back into speech. The result is presented to the user as a voice response.

[0411] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0412] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0413] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0414] [Third Embodiment]

[0415] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0416] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0417] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0418] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0419] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0420] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0421] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0422] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0423] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0424] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0425] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0426] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0427] This invention provides a system for streamlining customer complaint handling in call centers. This system combines speech recognition, natural language processing, and machine learning models, and operates in the following manner.

[0428] When a user calls a call center, the terminal sends the audio as digital audio data to the server. The server receives this audio data and converts it into text data using a speech recognition engine. The server then uses natural language processing techniques to analyze the text data in detail. The purpose of this analysis is to identify the types of complaints and emotional expressions contained in the text and to assess their severity based on these.

[0429] The analysis results are input into an AI model for handling complaints. The server utilizes the AI ​​model to generate the optimal response message based on similar past cases and statistical data. The response message is dynamically customized according to the type and severity of the complaint before being delivered to the user.

[0430] The terminal receives a response message sent from the server and conveys it to the user using speech synthesis technology. The content of the response varies widely, including confirmation of product description, apologies, and refund offers. If follow-up actions are to be taken, it may also include related details and schedules.

[0431] Furthermore, the server tracks the response results in real time, preparing for user reactions and any additional complaints. If the problem is not resolved, the server immediately sends an alert to the operator, requesting direct action.

[0432] As a concrete example, consider a case where a user complains about a delay in their order. After the terminal receives the audio, the server transcribes the audio into text and performs analysis to identify the cause of the complaint. The AI ​​model quickly identifies the keyword "delay" and, referencing effective responses in similar situations in the past, automatically generates a message explaining the reason for the delay and an apology. This ensures fast and consistent service, improving user satisfaction and reducing the burden on operators.

[0433] The following describes the processing flow.

[0434] Step 1:

[0435] A user makes a call to the call center, and the call begins. The device receives the call and records it as digital audio.

[0436] Step 2:

[0437] The terminal streams the recorded audio data to the server. The server stores the received audio data.

[0438] Step 3:

[0439] The server uses a speech recognition engine to convert the audio data into text data. This converts the audio content into a text format suitable for analysis.

[0440] Step 4:

[0441] The server applies natural language processing techniques to analyze the converted text data. The goal here is to identify the content of the complaint, the sentiment behind it, and important keywords.

[0442] Step 5:

[0443] Based on the analysis results, the server inputs the data into an AI model to determine the type and severity of the claim. This AI model compares the data with past data to recognize patterns.

[0444] Step 6:

[0445] The server generates the optimal response message for the user based on the AI ​​model's judgment. This message is dynamically assembled based on a predefined template.

[0446] Step 7:

[0447] The server sends the generated response message to the terminal. The terminal either reads this message aloud using speech synthesis technology or displays it as text.

[0448] Step 8:

[0449] The server monitors user responses, obtains follow-up actions as needed, and determines the execution plan.

[0450] Step 9:

[0451] If the server determines that a problem remains unresolved, or if an anomaly is detected in the system, it sends an alert to the operator. This allows the operator to intervene when necessary.

[0452] (Example 1)

[0453] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0454] Handling customer complaints in call centers is extremely diverse and requires both speed and accuracy, but traditional methods make it difficult to do so efficiently. Furthermore, it is necessary to reduce the burden on workers while simultaneously improving customer satisfaction. To address this challenge, there is a need for methods that automate speech conversion and analysis, as well as automate response generation.

[0455] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0456] In this invention, the server includes an information processing device for acquiring audio data, an acoustic analysis device for converting the acquired audio data into text data, and a language processing device for analyzing the text data and determining the type and importance of the information. This enables rapid and accurate analysis and automatic generation of responses to customer complaints.

[0457] An "information processing device for acquiring voice data" is a device that captures voice data from phone calls and other sources and converts it into a format usable within the system.

[0458] An "acoustic analysis device that converts acquired audio data into text data" is a device that uses speech recognition technology to perform the process of converting audio signals into text information.

[0459] A "language processing device that analyzes text data to determine the type and importance of information" is a device that utilizes natural language processing technology to extract and determine meaning, sentiment, and importance from text data.

[0460] A "machine learning device that generates appropriate response information based on judgment" is a device that creates the optimal response message according to the situation, based on past data and learning algorithms.

[0461] A "communication device that provides response information to a user" is a device that provides a means of transmitting generated response information to the user in voice or text format.

[0462] The "function to monitor the results of the response and send an alert to the worker if the problem is not resolved" means that the system tracks the processing results and user responses, and directly notifies a human if the problem remains unresolved under pre-set conditions.

[0463] This invention is a system primarily aimed at automating and streamlining customer complaint handling in call centers. The system processes voice data, analyzes text data using natural language processing technology, and provides the user with the most appropriate response. A detailed configuration is shown below.

[0464] The server acquires the voice spoken by the user through the terminal. The terminal converts the voice data into a digital format and transmits it to the server. The server uses speech recognition software (e.g., a general speech recognition engine) to convert this voice data into text data. The transcription of the voice provides the basic information necessary to accurately determine the type and severity of the claim.

[0465] This analysis uses specific natural language processing libraries (for example, open-source natural language processing libraries). In the sentiment analysis stage, sentiment keywords are extracted from the text to understand the user's emotional state. Furthermore, the server uses machine learning models to generate response messages appropriate for similar cases based on the data collected so far.

[0466] Generative AI models are used to analyze historical data and produce the most effective responses. For example, in response to a customer complaint about a delayed product, the AI ​​model appropriately constructs a message that includes an apology for the delay and details of the reason for the delay. This response is customized to the content of the complaint before being delivered to the user.

[0467] The terminal can use text-to-speech technology (e.g., a common speech synthesis system) to provide users with feedback in a natural voice based on text messages provided by the server. This process allows users to receive quicker complaint resolution, leading to increased satisfaction.

[0468] An example of a prompt might be, "Please provide information regarding the product delay and generate an apology voice message." Using this prompt, the system can demonstrate its ability to generate an appropriate response to a specific complaint.

[0469] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0470] Step 1:

[0471] When a user makes a phone call, the device receives the call and generates audio data in real time. The input is the user's voice, and the output is digital audio data. The specific action performed by the device is to collect the voice via the microphone and record it in digital format.

[0472] Step 2:

[0473] The terminal sends the generated audio data to the server. The input is the digital audio data acquired in the previous step, and the output is a compressed audio file sent to the server. The specific actions taken to reach the server involve the terminal compressing the data and sending it using a secure communication protocol.

[0474] Step 3:

[0475] The server uses speech recognition software to convert speech data into text data. The input is digital speech data received from the terminal, and the output is text data that transcribes the speech into written form. Specifically, the server applies a noise reduction filter and uses a language model to perform the process of converting speech to text.

[0476] Step 4:

[0477] The server analyzes the obtained text data using natural language processing techniques to determine the type and sentiment of the claim. The input is text data, and the output is an evaluation of the claim's attributes and sentiment. Specifically, the server performs grammatical structure analysis and key phrase extraction to identify important information within the text.

[0478] Step 5:

[0479] The server uses a generative AI model to generate the optimal response message based on the analysis results. The input is the analyzed claims attributes and emotions, and the output is the response text to be conveyed to the user. The server's specific operation is for the model to select a response template based on past data and incorporate it into specific wording.

[0480] Step 6:

[0481] The generated response message is sent to the terminal, which then uses speech synthesis technology to deliver the response message to the user verbally. The input is the response text from the server, and the output is the voice response delivered to the user. Specifically, the terminal uses an appropriate speech synthesis engine to play the message in a natural-sounding voice.

[0482] Step 7:

[0483] The server monitors user responses and evaluates the outcome of the response. Inputs are additional voice responses and evaluation feedback from the user, and outputs are decisions on the next action based on the results. Specifically, in this step, the server performs voice analysis in real time and activates an alarm system if the problem remains unresolved.

[0484] (Application Example 1)

[0485] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0486] In autonomously moving machines, it is essential to recognize user complaints and operational dissatisfaction in real time and respond quickly and accurately. Conventional systems have sometimes resulted in delays in complaint handling, degrading the user experience. Furthermore, it has been difficult to immediately propose improvements to the route and operation, hindering efficient operation. This invention solves these problems.

[0487] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0488] In this invention, the server includes means for acquiring voice data as a claim, means for converting the acquired voice data into text data, means for analyzing the text data to determine the classification and severity of the claim, means for dynamically generating an appropriate response message based on the determination, means for providing the generated response message by speech synthesis, and means for transmitting a command when immediate response by a human operator is required. This enables real-time handling of claims and suggestions for improving routes and operations.

[0489] "Audio data" refers to data that represents information generated as sound in a digital format, and is subject to analysis and conversion.

[0490] "Text data" refers to data obtained by converting audio data into string information, and is used as a foundation for natural language processing.

[0491] "Claim classification" is the process of identifying claims that belong to a specific type or category by analyzing text data.

[0492] "Severity" is a measure used to assess the magnitude and urgency of the impact a complaint has on business operations or services.

[0493] A "response message" is a message provided to the user that shows an appropriate reaction to a complaint.

[0494] "Speech synthesis" is a technology that outputs generated text-based response messages as speech that closely resembles a human voice.

[0495] A "human operator" is a human worker who intervenes and takes appropriate action when autonomous machine operation is difficult.

[0496] "Sending a command" refers to the act of notifying a human operator to take action when specific conditions occur.

[0497] This invention provides a system for autonomously moving machines to receive and immediately respond to user complaints during operation. An embodiment of this system is described below.

[0498] The server acquires audio data transmitted from vehicles and mobile devices. A speech recognition engine is used to convert this audio data into text data. Specifically, Google Speech-to-Text and Amazon Transcribe can be used as speech recognition APIs. The converted text data is analyzed using natural language processing libraries (e.g., spaCy and NLTK) to classify the claims and determine their severity.

[0499] Next, a generative AI model running on a cloud server dynamically generates the optimal response message based on the analysis results. This AI model learns from similar past claim data and presents an action plan in real time to resolve the user's problem. The generated response message is delivered to the user via voice from inside the vehicle or on a device using speech synthesis technology (e.g., Google Text-to-Speech).

[0500] Furthermore, the server can suggest improvements to the operating path and maneuvers of the autonomously moving machine. For example, if a user complains that "the brakes are too abrupt," the server can refer to past data and provide real-time feedback on improved braking operation.

[0501] Furthermore, if the problem is not resolved, it has the function to generate a warning and send a command to a human operator. This enables a quick response and improves the user experience.

[0502] Prompt example:

[0503] "What information is needed to analyze user comments and generate optimal feedback?"

[0504] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0505] Step 1:

[0506] The terminal acquires voice data from the user. The acquired voice data is converted into a digital format and sent to the server. This allows the voice information to be obtained in a digital format suitable for analysis.

[0507] Step 2:

[0508] The server converts the received audio data into text data using a speech recognition engine. For example, it uses Google Speech-to-Text to analyze the format of the audio waveform and convert it into text. This output text obtained from the input audio is then used for subsequent analysis.

[0509] Step 3:

[0510] The server analyzes text data using natural language processing techniques to classify and determine the severity of claims. It utilizes spaCy and NLTK to analyze the text content and extract specific keywords (e.g., sudden braking, delay). The analysis output based on the input text is the specific category of the claim and its evaluation.

[0511] Step 4:

[0512] Based on the analysis of the complaint, the server uses a generative AI model to generate the optimal response message. It refers to previously learned similar cases and statistical data to prepare a dynamic reply to the user. This process is text-based, and the AI ​​model generates an output message incorporating information obtained from the input analysis.

[0513] Step 5:

[0514] The server converts the generated response message into speech using a speech synthesis engine (e.g., Google Text-to-Speech) and provides it to the user via the device. By outputting text-based messages as speech and presenting them directly to the user, it enables easily understandable communication.

[0515] Step 6:

[0516] If the problem is not resolved, the server generates a warning and sends instructions to a human operator. If action is required, it generates an alert and sends a notification to the appropriate management system to enable immediate human intervention.

[0517] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0518] This invention provides a system for efficiently and effectively handling customer complaints in a call center. This system has an integrated configuration of speech recognition, natural language processing, an emotion engine, and a machine learning model.

[0519] When a user makes a call, the terminal receives the voice call and sends it to the server as digital audio data. The server converts the received audio data into text data using a speech recognition engine. The server then applies natural language processing techniques to this converted text data to analyze the type of claim, key keywords, and severity.

[0520] Furthermore, the server incorporates an emotion engine that can analyze user emotions from text data. The emotion engine analyzes the user's vocabulary and tone of voice to recognize emotions such as satisfaction, dissatisfaction, and anger in real time.

[0521] The analyzed data is processed by an AI model to dynamically generate response messages tailored to the user's complaint and emotional state. When these generated response messages are delivered to the user, they are adjusted according to their emotions, ensuring appropriate tone and content for effective communication.

[0522] For example, if a user expresses anger about a defective product, the device records the audio and sends it to a server. The server analyzes the audio, identifies the emotion of anger, and generates a message that includes an apology and a replacement offer. The device then uses speech synthesis to deliver this message to the user. The server also determines if follow-up is necessary and sends an alert to an operator to prompt a quick response.

[0523] The server continuously monitors the response results and improves and adjusts the response measures as needed. This system is expected to improve the quality of complaint handling and increase customer satisfaction.

[0524] The following describes the processing flow.

[0525] Step 1:

[0526] A user makes a call to a call center, and the device receives the voice call. The call content is transmitted to the server in real time as voice data.

[0527] Step 2:

[0528] The server activates the speech recognition engine and converts the received audio data into text data. This makes the audio content available in text format.

[0529] Step 3:

[0530] The server analyzes the text data using natural language processing. The analysis extracts the type of complaint and keywords, while the sentiment engine simultaneously identifies the user's emotions.

[0531] Step 4:

[0532] The emotion engine evaluates the user's emotional state (e.g., anger, dissatisfaction, satisfaction) in real time based on the tone and word choice of their speech. Data including an emotion score is then generated.

[0533] Step 5:

[0534] The server inputs the analysis results and emotional data into the AI ​​model. The AI ​​model generates the optimal response message based on the characteristics of the complaint and the user's emotional state.

[0535] Step 6:

[0536] The server sends the generated response message to the terminal. The terminal uses speech synthesis technology to provide the user with the response message in either voice or text format.

[0537] Step 7:

[0538] The server monitors user responses and determines whether follow-up action is needed. If necessary, it sends a notification to the operator.

[0539] Step 8:

[0540] If the user submits another complaint or feedback, the server will initiate a new analysis and repeat the process. This process will continue until the user's problem is resolved and they are satisfied.

[0541] (Example 2)

[0542] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0543] Handling complaints in a call center presents a challenging task, requiring efficient and effective responses that are sensitive to the user's emotions. It is also crucial to establish an appropriate tracking system to quickly resolve user dissatisfaction. Furthermore, accurately monitoring the outcome of responses and taking immediate action if the problem remains unresolved is essential.

[0544] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0545] In this invention, the server includes means for acquiring voice information, means for converting the acquired voice information into text information, means for analyzing the text information to determine the type and severity of the claim, means for adjusting the generated response message to match the user's emotions, means for providing the adjusted response message by speech synthesis, means for determining a tracking function for the user based on the type and severity of the claim and the user's emotions, and further means for monitoring the processing results and sending a notification to the person in charge if the problem is not resolved. This enables appropriate responses that are in line with the user's emotions and rapid problem resolution in claim handling.

[0546] "Voice information" refers to information, including the speaker's voice, obtained through telephones or communication systems.

[0547] "Textual information" refers to information in a readable string format, obtained by converting audio information using speech recognition technology.

[0548] "Complaint type" refers to a category used to classify the content of user dissatisfaction or requests, which is analyzed from the acquired textual information.

[0549] "Severity" refers to the degree of urgency or importance that requires action, as determined based on the type of claim.

[0550] A "response message" is the content of the reply provided to the user based on the results of analyzing voice and text information.

[0551] "Speech synthesis" is a technology that mechanically generates speech based on textual information and provides information to the user audibly.

[0552] The "tracking function" is a feature that monitors the progress of handling complaints and prompts further action as needed.

[0553] A "notification" is a means of communication used to warn the person in charge if the problem is not resolved based on the results of the response.

[0554] This invention provides a system for efficiently and effectively handling customer complaints in a call center. This system integrates speech recognition technology, natural language processing technology, sentiment analysis technology, and machine learning models.

[0555] When a user makes a phone call, the device receives the voice call, converts the voice data into a digital format, and sends it to the server. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice data into text data. The converted text data is then analyzed using natural language processing techniques (e.g., spaCy or NLTK) to identify the type of claim, key keywords, and severity.

[0556] The server incorporates an emotion engine (e.g., IBM Watson Tone Analyzer) that analyzes user emotions from text data. By analyzing the user's vocabulary and tone of voice, emotions such as satisfaction, dissatisfaction, and anger are recognized in real time. The analyzed data is processed by an AI model (e.g., OpenAI GPT-3) to dynamically generate response messages tailored to the user's complaint and emotional state. These response messages are adjusted according to the emotion and delivered to the user in an appropriate tone and content.

[0557] For example, if a user expresses anger about a defective product, the device records the audio and sends it to a server. The server analyzes the audio, identifies the level of anger, and generates a message that includes an apology and a replacement offer. The device then delivers this message to the user using speech synthesis (e.g., Amazon Polly). The server can also determine if follow-up is necessary and send an alert to an operator to encourage a quick response.

[0558] An example of a prompt is, "Generate the optimal response based on the user's emotions and complaint details." This prompt prompts the AI ​​model to generate the most appropriate response message. This is expected to improve the quality of complaint handling and increase customer satisfaction.

[0559] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0560] Step 1:

[0561] The user makes a phone call. The terminal receives the voice call and converts the voice data into a digital format. The input is the user's voice, and the output is digital voice data. The terminal sends this digital voice data to the server.

[0562] Step 2:

[0563] The server inputs the received digital audio data into the speech recognition engine and converts it into text data. Here, the input is digital audio data, and the output is text data. The server then verifies whether the conversion was successful.

[0564] Step 3:

[0565] The server inputs the converted text data into a natural language processing engine. This engine analyzes the text data to identify the type of claim, key keywords, and severity. The input is text data, and the output is the analyzed category, keyword, and severity information.

[0566] Step 4:

[0567] The server inputs text data into its emotion analysis engine to analyze the user's emotions. The input is the text data obtained in step 3, and the output is information about emotions. Specifically, emotions such as satisfaction, dissatisfaction, and anger are identified.

[0568] Step 5:

[0569] The server inputs the analysis results into a generation AI model and generates a response message tailored to the user's complaint and emotional state. The input includes information on category, keywords, severity, and emotion, while the output is the response message.

[0570] Step 6:

[0571] The server adjusts the response message to match the user's emotions. In this step, the generated message and emotional information are used as input to output an optimized response message. Specifically, the tone and content of the message are adjusted.

[0572] Step 7:

[0573] The terminal inputs a pre-arranged response message into a speech synthesis engine, converts it into speech format, and delivers it to the user. The input is an optimized text message, and the output is synthesized speech. The terminal then plays this speech message to the user.

[0574] Step 8:

[0575] The server continuously monitors the response results. If the problem is not resolved, it sends a notification to the responsible person as needed to prompt action. In this step, it receives the processing results as input and outputs an alert if action is required.

[0576] (Application Example 2)

[0577] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0578] In voice-based request processing, it is essential to accurately assess the user's emotions and urgency, and to respond quickly and appropriately. However, conventional systems often fail to adequately analyze emotions or assess urgency, leading to decreased customer satisfaction. This is especially true in areas where security is paramount, requiring rapid risk prediction and appropriate response. Therefore, there is a need for technology that can highly analyze voice information and enable appropriate responses and follow-up.

[0579] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0580] In this invention, the server includes a device for acquiring voice information, a device for converting the acquired voice information into text information, a device for analyzing the text information to determine the type and severity of the request, and a device for analyzing emotions to determine the urgency. This enables a quick and appropriate response based on the user's emotions and urgency.

[0581] "Audio information" refers to data based on sound waves acquired from an input device.

[0582] "Textual information" refers to string data that is obtained by converting audio information through a recognition device.

[0583] A "requirement" is a concept that describes the problems or needs that a user expresses to a system.

[0584] "Severity" is a measure that indicates the level of importance and urgency of a request.

[0585] "Emotions" are elements that express the user's subjective mental state.

[0586] "Urgency" is a measure that indicates whether an immediate response is necessary to a particular situation.

[0587] A "device" is a collection of hardware and software used to process, analyze, and interpret audio and text information.

[0588] The system implementing this invention is designed to process requests using voice. When a user inputs voice information, the system's terminal collects that voice information and transmits it to a server. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice information into text information. This text information is then analyzed by a natural language processing engine (e.g., Google Natural Language API).

[0589] Based on the analyzed information, the server determines the type and severity of the request. In addition, a sentiment analysis API (e.g., IBM Watson Tone Analyzer) is used to identify the user's emotions and their urgency. This data is input into a generative AI model configured for automated response generation, which then generates appropriate response information. The generated response information is then converted back into speech using a speech synthesis engine (e.g., Amazon Polly) and provided to the user.

[0590] This system allows users to, for example, voice-input a statement like, "I've recently been feeling unsafe," and the system analyzes their feelings of anxiety. If it determines that a security patrol is necessary, measures are immediately taken. By taking appropriate follow-up actions according to the user's situation, improved user safety can be expected.

[0591] A concrete example of a prompt sentence corresponding to the generative AI model is, "Generate the most appropriate response based on the user's voice." This invention contributes to providing a quick response and a sense of security.

[0592] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0593] Step 1:

[0594] The device acquires voice information from the user. Voice information is collected when the user speaks into the smartphone, and this voice is stored digitally by the device. This voice information becomes the initial input data.

[0595] Step 2:

[0596] The device sends the acquired audio information to the server. The audio data is uploaded to the server via the internet. This transfer to the server enables the start of the next process.

[0597] Step 3:

[0598] The server uses a speech recognition engine to convert speech information into text. Specifically, it utilizes the Google Cloud Speech-to-Text API to convert speech data into text data. This process records the information obtained from the speech as text.

[0599] Step 4:

[0600] The server analyzes the text information using a natural language processing engine. Through the Google Natural Language API, it identifies the type and severity of the request from the text information. It then extracts important keywords from the input text data and performs a process to understand its context.

[0601] Step 5:

[0602] The server uses a sentiment analysis API to analyze the sentiment of text information. IBM Watson Tone Analyzer is used to determine the user's emotional state. Emotions are classified based on the phrasing and tone of the text data.

[0603] Step 6:

[0604] The server generates appropriate response information using a generated AI model. Based on the type, severity, and emotional state of the request obtained in the previous step, the AI ​​generates the optimal message using data including prompt text as input.

[0605] Step 7:

[0606] The server converts the generated response information into speech using a speech synthesis engine. Amazon Polly is then used to convert the generated text response back into speech. The result is presented to the user as a voice response.

[0607] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0608] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0609] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0610] [Fourth Embodiment]

[0611] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0612] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0613] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0614] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0615] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0616] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0617] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0618] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0619] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0620] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0621] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0622] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0623] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0624] This invention provides a system for streamlining customer complaint handling in call centers. This system combines speech recognition, natural language processing, and machine learning models, and operates in the following manner.

[0625] When a user calls a call center, the terminal sends the audio as digital audio data to the server. The server receives this audio data and converts it into text data using a speech recognition engine. The server then uses natural language processing techniques to analyze the text data in detail. The purpose of this analysis is to identify the types of complaints and emotional expressions contained in the text and to assess their severity based on these.

[0626] The analysis results are input into an AI model for handling complaints. The server utilizes the AI ​​model to generate the optimal response message based on similar past cases and statistical data. The response message is dynamically customized according to the type and severity of the complaint before being delivered to the user.

[0627] The terminal receives a response message sent from the server and conveys it to the user using speech synthesis technology. The content of the response varies widely, including confirmation of product description, apologies, and refund offers. If follow-up actions are to be taken, it may also include related details and schedules.

[0628] Furthermore, the server tracks the response results in real time, preparing for user reactions and any additional complaints. If the problem is not resolved, the server immediately sends an alert to the operator, requesting direct action.

[0629] As a concrete example, consider a case where a user complains about a delay in their order. After the terminal receives the audio, the server transcribes the audio into text and performs analysis to identify the cause of the complaint. The AI ​​model quickly identifies the keyword "delay" and, referencing effective responses in similar situations in the past, automatically generates a message explaining the reason for the delay and an apology. This ensures fast and consistent service, improving user satisfaction and reducing the burden on operators.

[0630] The following describes the processing flow.

[0631] Step 1:

[0632] A user makes a call to the call center, and the call begins. The device receives the call and records it as digital audio.

[0633] Step 2:

[0634] The terminal streams the recorded audio data to the server. The server stores the received audio data.

[0635] Step 3:

[0636] The server uses a speech recognition engine to convert the audio data into text data. This converts the audio content into a text format suitable for analysis.

[0637] Step 4:

[0638] The server applies natural language processing techniques to analyze the converted text data. The goal here is to identify the content of the complaint, the sentiment behind it, and important keywords.

[0639] Step 5:

[0640] Based on the analysis results, the server inputs the data into an AI model to determine the type and severity of the claim. This AI model compares the data with past data to recognize patterns.

[0641] Step 6:

[0642] The server generates the optimal response message for the user based on the AI ​​model's judgment. This message is dynamically assembled based on a predefined template.

[0643] Step 7:

[0644] The server sends the generated response message to the terminal. The terminal either reads this message aloud using speech synthesis technology or displays it as text.

[0645] Step 8:

[0646] The server monitors user responses, obtains follow-up actions as needed, and determines the execution plan.

[0647] Step 9:

[0648] If the server determines that a problem remains unresolved, or if an anomaly is detected in the system, it sends an alert to the operator. This allows the operator to intervene when necessary.

[0649] (Example 1)

[0650] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0651] Handling customer complaints in call centers is extremely diverse and requires both speed and accuracy, but traditional methods make it difficult to do so efficiently. Furthermore, it is necessary to reduce the burden on workers while simultaneously improving customer satisfaction. To address this challenge, there is a need for methods that automate speech conversion and analysis, as well as automate response generation.

[0652] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0653] In this invention, the server includes an information processing device for acquiring audio data, an acoustic analysis device for converting the acquired audio data into text data, and a language processing device for analyzing the text data and determining the type and importance of the information. This enables rapid and accurate analysis and automatic generation of responses to customer complaints.

[0654] An "information processing device for acquiring voice data" is a device that captures voice data from phone calls and other sources and converts it into a format usable within the system.

[0655] An "acoustic analysis device that converts acquired audio data into text data" is a device that uses speech recognition technology to perform the process of converting audio signals into text information.

[0656] A "language processing device that analyzes text data to determine the type and importance of information" is a device that utilizes natural language processing technology to extract and determine meaning, sentiment, and importance from text data.

[0657] A "machine learning device that generates appropriate response information based on judgment" is a device that creates the optimal response message according to the situation, based on past data and learning algorithms.

[0658] A "communication device that provides response information to a user" is a device that provides a means of transmitting generated response information to the user in voice or text format.

[0659] The "function to monitor the results of the response and send an alert to the worker if the problem is not resolved" means that the system tracks the processing results and user responses, and directly notifies a human if the problem remains unresolved under pre-set conditions.

[0660] This invention is a system primarily aimed at automating and streamlining customer complaint handling in call centers. The system processes voice data, analyzes text data using natural language processing technology, and provides the user with the most appropriate response. A detailed configuration is shown below.

[0661] The server acquires the voice spoken by the user through the terminal. The terminal converts the voice data into a digital format and transmits it to the server. The server uses speech recognition software (e.g., a general speech recognition engine) to convert this voice data into text data. The transcription of the voice provides the basic information necessary to accurately determine the type and severity of the claim.

[0662] This analysis uses specific natural language processing libraries (for example, open-source natural language processing libraries). In the sentiment analysis stage, sentiment keywords are extracted from the text to understand the user's emotional state. Furthermore, the server uses machine learning models to generate response messages appropriate for similar cases based on the data collected so far.

[0663] Generative AI models are used to analyze historical data and produce the most effective responses. For example, in response to a customer complaint about a delayed product, the AI ​​model appropriately constructs a message that includes an apology for the delay and details of the reason for the delay. This response is customized to the content of the complaint before being delivered to the user.

[0664] The terminal can use text-to-speech technology (e.g., a common speech synthesis system) to provide users with feedback in a natural voice based on text messages provided by the server. This process allows users to receive quicker complaint resolution, leading to increased satisfaction.

[0665] An example of a prompt might be, "Please provide information regarding the product delay and generate an apology voice message." Using this prompt, the system can demonstrate its ability to generate an appropriate response to a specific complaint.

[0666] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0667] Step 1:

[0668] When a user makes a phone call, the device receives the call and generates audio data in real time. The input is the user's voice, and the output is digital audio data. The specific action performed by the device is to collect the voice via the microphone and record it in digital format.

[0669] Step 2:

[0670] The terminal sends the generated audio data to the server. The input is the digital audio data acquired in the previous step, and the output is a compressed audio file sent to the server. The specific actions taken to reach the server involve the terminal compressing the data and sending it using a secure communication protocol.

[0671] Step 3:

[0672] The server uses speech recognition software to convert speech data into text data. The input is digital speech data received from the terminal, and the output is text data that transcribes the speech into written form. Specifically, the server applies a noise reduction filter and uses a language model to perform the process of converting speech to text.

[0673] Step 4:

[0674] The server analyzes the obtained text data using natural language processing techniques to determine the type and sentiment of the claim. The input is text data, and the output is an evaluation of the claim's attributes and sentiment. Specifically, the server performs grammatical structure analysis and key phrase extraction to identify important information within the text.

[0675] Step 5:

[0676] The server uses a generative AI model to generate the optimal response message based on the analysis results. The input is the analyzed claims attributes and emotions, and the output is the response text to be conveyed to the user. The server's specific operation is for the model to select a response template based on past data and incorporate it into specific wording.

[0677] Step 6:

[0678] The generated response message is sent to the terminal, which then uses speech synthesis technology to deliver the response message to the user verbally. The input is the response text from the server, and the output is the voice response delivered to the user. Specifically, the terminal uses an appropriate speech synthesis engine to play the message in a natural-sounding voice.

[0679] Step 7:

[0680] The server monitors user responses and evaluates the outcome of the response. Inputs are additional voice responses and evaluation feedback from the user, and outputs are decisions on the next action based on the results. Specifically, in this step, the server performs voice analysis in real time and activates an alarm system if the problem remains unresolved.

[0681] (Application Example 1)

[0682] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0683] In autonomously moving machines, it is essential to recognize user complaints and operational dissatisfaction in real time and respond quickly and accurately. Conventional systems have sometimes resulted in delays in complaint handling, degrading the user experience. Furthermore, it has been difficult to immediately propose improvements to the route and operation, hindering efficient operation. This invention solves these problems.

[0684] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0685] In this invention, the server includes means for acquiring voice data as a claim, means for converting the acquired voice data into text data, means for analyzing the text data to determine the classification and severity of the claim, means for dynamically generating an appropriate response message based on the determination, means for providing the generated response message by speech synthesis, and means for transmitting a command when immediate response by a human operator is required. This enables real-time handling of claims and suggestions for improving routes and operations.

[0686] "Audio data" refers to data that represents information generated as sound in a digital format, and is subject to analysis and conversion.

[0687] "Text data" refers to data obtained by converting audio data into string information, and is used as a foundation for natural language processing.

[0688] "Claim classification" is the process of identifying claims that belong to a specific type or category by analyzing text data.

[0689] "Severity" is a measure used to assess the magnitude and urgency of the impact a complaint has on business operations or services.

[0690] A "response message" is a message provided to the user that shows an appropriate reaction to a complaint.

[0691] "Speech synthesis" is a technology that outputs generated text-based response messages as speech that closely resembles a human voice.

[0692] A "human operator" is a human worker who intervenes and takes appropriate action when autonomous machine operation is difficult.

[0693] "Sending a command" refers to the act of notifying a human operator to take action when specific conditions occur.

[0694] This invention provides a system for autonomously moving machines to receive and immediately respond to user complaints during operation. An embodiment of this system is described below.

[0695] The server acquires audio data transmitted from vehicles and mobile devices. A speech recognition engine is used to convert this audio data into text data. Specifically, Google Speech-to-Text and Amazon Transcribe can be used as speech recognition APIs. The converted text data is analyzed using natural language processing libraries (e.g., spaCy and NLTK) to classify the claims and determine their severity.

[0696] Next, a generative AI model running on a cloud server dynamically generates the optimal response message based on the analysis results. This AI model learns from similar past claim data and presents an action plan in real time to resolve the user's problem. The generated response message is delivered to the user via voice from inside the vehicle or on a device using speech synthesis technology (e.g., Google Text-to-Speech).

[0697] Furthermore, the server can suggest improvements to the operating path and maneuvers of the autonomously moving machine. For example, if a user complains that "the brakes are too abrupt," the server can refer to past data and provide real-time feedback on improved braking operation.

[0698] Furthermore, if the problem is not resolved, it has the function to generate a warning and send a command to a human operator. This enables a quick response and improves the user experience.

[0699] Prompt example:

[0700] "What information is needed to analyze user comments and generate optimal feedback?"

[0701] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0702] Step 1:

[0703] The terminal acquires voice data from the user. The acquired voice data is converted into a digital format and sent to the server. This allows the voice information to be obtained in a digital format suitable for analysis.

[0704] Step 2:

[0705] The server converts the received audio data into text data using a speech recognition engine. For example, it uses Google Speech-to-Text to analyze the format of the audio waveform and convert it into text. This output text obtained from the input audio is then used for subsequent analysis.

[0706] Step 3:

[0707] The server analyzes text data using natural language processing techniques to classify and determine the severity of claims. It utilizes spaCy and NLTK to analyze the text content and extract specific keywords (e.g., sudden braking, delay). The analysis output based on the input text is the specific category of the claim and its evaluation.

[0708] Step 4:

[0709] Based on the analysis of the complaint, the server uses a generative AI model to generate the optimal response message. It refers to previously learned similar cases and statistical data to prepare a dynamic reply to the user. This process is text-based, and the AI ​​model generates an output message incorporating information obtained from the input analysis.

[0710] Step 5:

[0711] The server converts the generated response message into speech using a speech synthesis engine (e.g., Google Text-to-Speech) and provides it to the user via the device. By outputting text-based messages as speech and presenting them directly to the user, it enables easily understandable communication.

[0712] Step 6:

[0713] If the problem is not resolved, the server generates a warning and sends instructions to a human operator. If action is required, it generates an alert and sends a notification to the appropriate management system to enable immediate human intervention.

[0714] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0715] This invention provides a system for efficiently and effectively handling customer complaints in a call center. This system has an integrated configuration of speech recognition, natural language processing, an emotion engine, and a machine learning model.

[0716] When a user makes a call, the terminal receives the voice call and sends it to the server as digital audio data. The server converts the received audio data into text data using a speech recognition engine. The server then applies natural language processing techniques to this converted text data to analyze the type of claim, key keywords, and severity.

[0717] Furthermore, the server incorporates an emotion engine that can analyze user emotions from text data. The emotion engine analyzes the user's vocabulary and tone of voice to recognize emotions such as satisfaction, dissatisfaction, and anger in real time.

[0718] The analyzed data is processed by an AI model to dynamically generate response messages tailored to the user's complaint and emotional state. When these generated response messages are delivered to the user, they are adjusted according to their emotions, ensuring appropriate tone and content for effective communication.

[0719] For example, if a user expresses anger about a defective product, the device records the audio and sends it to a server. The server analyzes the audio, identifies the emotion of anger, and generates a message that includes an apology and a replacement offer. The device then uses speech synthesis to deliver this message to the user. The server also determines if follow-up is necessary and sends an alert to an operator to prompt a quick response.

[0720] The server continuously monitors the response results and improves and adjusts the response measures as needed. This system is expected to improve the quality of complaint handling and increase customer satisfaction.

[0721] The following describes the processing flow.

[0722] Step 1:

[0723] A user makes a call to a call center, and the device receives the voice call. The call content is transmitted to the server in real time as voice data.

[0724] Step 2:

[0725] The server activates the speech recognition engine and converts the received audio data into text data. This makes the audio content available in text format.

[0726] Step 3:

[0727] The server analyzes the text data using natural language processing. The analysis extracts the type of complaint and keywords, while the sentiment engine simultaneously identifies the user's emotions.

[0728] Step 4:

[0729] The emotion engine evaluates the user's emotional state (e.g., anger, dissatisfaction, satisfaction) in real time based on the tone and word choice of their speech. Data including an emotion score is then generated.

[0730] Step 5:

[0731] The server inputs the analysis results and emotional data into the AI ​​model. The AI ​​model generates the optimal response message based on the characteristics of the complaint and the user's emotional state.

[0732] Step 6:

[0733] The server sends the generated response message to the terminal. The terminal uses speech synthesis technology to provide the user with the response message in either voice or text format.

[0734] Step 7:

[0735] The server monitors user responses and determines whether follow-up action is needed. If necessary, it sends a notification to the operator.

[0736] Step 8:

[0737] If the user submits another complaint or feedback, the server will initiate a new analysis and repeat the process. This process will continue until the user's problem is resolved and they are satisfied.

[0738] (Example 2)

[0739] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0740] Handling complaints in a call center presents a challenging task, requiring efficient and effective responses that are sensitive to the user's emotions. It is also crucial to establish an appropriate tracking system to quickly resolve user dissatisfaction. Furthermore, accurately monitoring the outcome of responses and taking immediate action if the problem remains unresolved is essential.

[0741] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0742] In this invention, the server includes means for acquiring voice information, means for converting the acquired voice information into text information, means for analyzing the text information to determine the type and severity of the claim, means for adjusting the generated response message to match the user's emotions, means for providing the adjusted response message by speech synthesis, means for determining a tracking function for the user based on the type and severity of the claim and the user's emotions, and further means for monitoring the processing results and sending a notification to the person in charge if the problem is not resolved. This enables appropriate responses that are in line with the user's emotions and rapid problem resolution in claim handling.

[0743] "Voice information" refers to information, including the speaker's voice, obtained through telephones or communication systems.

[0744] "Textual information" refers to information in a readable string format, obtained by converting audio information using speech recognition technology.

[0745] "Complaint type" refers to a category used to classify the content of user dissatisfaction or requests, which is analyzed from the acquired textual information.

[0746] "Severity" refers to the degree of urgency or importance that requires action, as determined based on the type of claim.

[0747] A "response message" is the content of the reply provided to the user based on the results of analyzing voice and text information.

[0748] "Speech synthesis" is a technology that mechanically generates speech based on textual information and provides information to the user audibly.

[0749] The "tracking function" is a feature that monitors the progress of handling complaints and prompts further action as needed.

[0750] A "notification" is a means of communication used to warn the person in charge if the problem is not resolved based on the results of the response.

[0751] This invention provides a system for efficiently and effectively handling customer complaints in a call center. This system integrates speech recognition technology, natural language processing technology, sentiment analysis technology, and machine learning models.

[0752] When a user makes a phone call, the device receives the voice call, converts the voice data into a digital format, and sends it to the server. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice data into text data. The converted text data is then analyzed using natural language processing techniques (e.g., spaCy or NLTK) to identify the type of claim, key keywords, and severity.

[0753] The server incorporates an emotion engine (e.g., IBM Watson Tone Analyzer) that analyzes user emotions from text data. By analyzing the user's vocabulary and tone of voice, emotions such as satisfaction, dissatisfaction, and anger are recognized in real time. The analyzed data is processed by an AI model (e.g., OpenAI GPT-3) to dynamically generate response messages tailored to the user's complaint and emotional state. These response messages are adjusted according to the emotion and delivered to the user in an appropriate tone and content.

[0754] For example, if a user expresses anger about a defective product, the device records the audio and sends it to a server. The server analyzes the audio, identifies the level of anger, and generates a message that includes an apology and a replacement offer. The device then delivers this message to the user using speech synthesis (e.g., Amazon Polly). The server can also determine if follow-up is necessary and send an alert to an operator to encourage a quick response.

[0755] An example of a prompt is, "Generate the optimal response based on the user's emotions and complaint details." This prompt prompts the AI ​​model to generate the most appropriate response message. This is expected to improve the quality of complaint handling and increase customer satisfaction.

[0756] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0757] Step 1:

[0758] The user makes a phone call. The terminal receives the voice call and converts the voice data into a digital format. The input is the user's voice, and the output is digital voice data. The terminal sends this digital voice data to the server.

[0759] Step 2:

[0760] The server inputs the received digital audio data into the speech recognition engine and converts it into text data. Here, the input is digital audio data, and the output is text data. The server then verifies whether the conversion was successful.

[0761] Step 3:

[0762] The server inputs the converted text data into a natural language processing engine. This engine analyzes the text data to identify the type of claim, key keywords, and severity. The input is text data, and the output is the analyzed category, keyword, and severity information.

[0763] Step 4:

[0764] The server inputs text data into its emotion analysis engine to analyze the user's emotions. The input is the text data obtained in step 3, and the output is information about emotions. Specifically, emotions such as satisfaction, dissatisfaction, and anger are identified.

[0765] Step 5:

[0766] The server inputs the analysis results into a generation AI model and generates a response message tailored to the user's complaint and emotional state. The input includes information on category, keywords, severity, and emotion, while the output is the response message.

[0767] Step 6:

[0768] The server adjusts the response message to match the user's emotions. In this step, the generated message and emotional information are used as input to output an optimized response message. Specifically, the tone and content of the message are adjusted.

[0769] Step 7:

[0770] The terminal inputs a pre-arranged response message into a speech synthesis engine, converts it into speech format, and delivers it to the user. The input is an optimized text message, and the output is synthesized speech. The terminal then plays this speech message to the user.

[0771] Step 8:

[0772] The server continuously monitors the response results. If the problem is not resolved, it sends a notification to the responsible person as needed to prompt action. In this step, it receives the processing results as input and outputs an alert if action is required.

[0773] (Application Example 2)

[0774] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0775] In voice-based request processing, it is essential to accurately assess the user's emotions and urgency, and to respond quickly and appropriately. However, conventional systems often fail to adequately analyze emotions or assess urgency, leading to decreased customer satisfaction. This is especially true in areas where security is paramount, requiring rapid risk prediction and appropriate response. Therefore, there is a need for technology that can highly analyze voice information and enable appropriate responses and follow-up.

[0776] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0777] In this invention, the server includes a device for acquiring voice information, a device for converting the acquired voice information into text information, a device for analyzing the text information to determine the type and severity of the request, and a device for analyzing emotions to determine the urgency. This enables a quick and appropriate response based on the user's emotions and urgency.

[0778] "Audio information" refers to data based on sound waves acquired from an input device.

[0779] "Textual information" refers to string data that is obtained by converting audio information through a recognition device.

[0780] A "requirement" is a concept that describes the problems or needs that a user expresses to a system.

[0781] "Severity" is a measure that indicates the level of importance and urgency of a request.

[0782] "Emotions" are elements that express the user's subjective mental state.

[0783] "Urgency" is a measure that indicates whether an immediate response is necessary to a particular situation.

[0784] A "device" is a collection of hardware and software used to process, analyze, and interpret audio and text information.

[0785] The system implementing this invention is designed to process requests using voice. When a user inputs voice information, the system's terminal collects that voice information and transmits it to a server. The server uses a speech recognition engine (e.g., Google Cloud Speech-to-Text API) to convert the voice information into text information. This text information is then analyzed by a natural language processing engine (e.g., Google Natural Language API).

[0786] Based on the analyzed information, the server determines the type and severity of the request. In addition, a sentiment analysis API (e.g., IBM Watson Tone Analyzer) is used to identify the user's emotions and their urgency. This data is input into a generative AI model configured for automated response generation, which then generates appropriate response information. The generated response information is then converted back into speech using a speech synthesis engine (e.g., Amazon Polly) and provided to the user.

[0787] This system allows users to, for example, voice-input a statement like, "I've recently been feeling unsafe," and the system analyzes their feelings of anxiety. If it determines that a security patrol is necessary, measures are immediately taken. By taking appropriate follow-up actions according to the user's situation, improved user safety can be expected.

[0788] A concrete example of a prompt sentence corresponding to the generative AI model is, "Generate the most appropriate response based on the user's voice." This invention contributes to providing a quick response and a sense of security.

[0789] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0790] Step 1:

[0791] The device acquires voice information from the user. Voice information is collected when the user speaks into the smartphone, and this voice is stored digitally by the device. This voice information becomes the initial input data.

[0792] Step 2:

[0793] The device sends the acquired audio information to the server. The audio data is uploaded to the server via the internet. This transfer to the server enables the start of the next process.

[0794] Step 3:

[0795] The server uses a speech recognition engine to convert speech information into text. Specifically, it utilizes the Google Cloud Speech-to-Text API to convert speech data into text data. This process records the information obtained from the speech as text.

[0796] Step 4:

[0797] The server analyzes the text information using a natural language processing engine. Through the Google Natural Language API, it identifies the type and severity of the request from the text information. It then extracts important keywords from the input text data and performs a process to understand its context.

[0798] Step 5:

[0799] The server uses a sentiment analysis API to analyze the sentiment of text information. IBM Watson Tone Analyzer is used to determine the user's emotional state. Emotions are classified based on the phrasing and tone of the text data.

[0800] Step 6:

[0801] The server generates appropriate response information using a generated AI model. Based on the type, severity, and emotional state of the request obtained in the previous step, the AI ​​generates the optimal message using data including prompt text as input.

[0802] Step 7:

[0803] The server converts the generated response information into speech using a speech synthesis engine. Amazon Polly is then used to convert the generated text response back into speech. The result is presented to the user as a voice response.

[0804] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0805] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0806] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0807] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0808] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0809] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0810] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0811] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0812] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0813] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0814] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0815] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0816] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0817] 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.

[0818] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0819] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0820] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0821] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0822] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0823] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0824] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0825] The following is further disclosed regarding the embodiments described above.

[0826] (Claim 1)

[0827] Means for obtaining audio data as a complaint,

[0828] A means of converting acquired audio data into text data,

[0829] A means of analyzing text data to determine the type and severity of a claim,

[0830] Means for generating an appropriate response message based on judgment,

[0831] A means of providing a response message to the user,

[0832] A system that includes this.

[0833] (Claim 2)

[0834] The system according to claim 1, further comprising means for determining follow-up actions for the user based on the type and severity of the claim.

[0835] (Claim 3)

[0836] The system according to claim 1, further comprising means for monitoring the response results and sending an alert to the operator if the problem is not resolved.

[0837] "Example 1"

[0838] (Claim 1)

[0839] An information processing device that acquires audio data,

[0840] An acoustic analysis device that converts acquired audio data into text data,

[0841] A language processing device that analyzes text data to determine the type and importance of the information,

[0842] A machine learning device that generates appropriate response information based on judgment,

[0843] A communication device that provides response information to the user,

[0844] A system that includes this.

[0845] (Claim 2)

[0846] The system according to claim 1, further comprising a function for determining ongoing countermeasures for the user based on the type and importance of the information.

[0847] (Claim 3)

[0848] The system according to claim 1, further comprising a function to monitor the results of the response and send an alert to the worker if the problem is not resolved.

[0849] "Application Example 1"

[0850] (Claim 1)

[0851] Means for obtaining audio data as a complaint,

[0852] A means of converting acquired audio data into text data,

[0853] A means of analyzing text data to classify and determine the severity of claims,

[0854] A means for dynamically generating an appropriate response message based on a judgment,

[0855] A means for providing the generated response message by speech synthesis,

[0856] A means of transmitting commands when immediate response by a human operator is required,

[0857] A system that includes this.

[0858] (Claim 2)

[0859] The system according to claim 1, further comprising means for suggesting improvements to the route and operation of a machine in autonomous motion based on the classification and severity of the claims.

[0860] (Claim 3)

[0861] The system according to claim 1, further comprising means for monitoring the response results, generating a warning if the issue is not resolved, and notifying a human operator.

[0862] "Example 2 of combining an emotion engine"

[0863] (Claim 1)

[0864] Means for acquiring audio information,

[0865] A means of converting acquired audio information into text information,

[0866] A means of analyzing textual information to determine the type and severity of a claim,

[0867] Means for generating an appropriate response message based on judgment,

[0868] A means of adjusting the generated response message to match the user's emotions,

[0869] A means for providing a pre-arranged response message by speech synthesis,

[0870] A system that includes this.

[0871] (Claim 2)

[0872] The system according to claim 1, further comprising means for determining tracking capabilities for a user based on the type and severity of the claim and the user's sentiment.

[0873] (Claim 3)

[0874] The system according to claim 1, further comprising means for monitoring processing results and sending a notification to the person in charge if the problem is not resolved.

[0875] "Application example 2 when combining with an emotional engine"

[0876] (Claim 1)

[0877] A device for acquiring audio information,

[0878] A device that converts acquired audio information into text information,

[0879] A device that analyzes textual information to determine the type and severity of a request,

[0880] A device that generates appropriate response information based on judgment,

[0881] A device that provides response information to the user,

[0882] A device that analyzes emotions and determines the level of urgency,

[0883] A system that includes this.

[0884] (Claim 2)

[0885] The system according to claim 1, further comprising a device for analyzing user emotions and determining follow-up actions based on urgency.

[0886] (Claim 3)

[0887] The system according to claim 1, further comprising a device for continuously monitoring the results of the response and sending a warning to a professional if the problem is not resolved. [Explanation of symbols]

[0888] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. Means for obtaining audio data as a complaint, A means of converting acquired audio data into text data, A means of analyzing text data to classify and determine the severity of claims, A means for dynamically generating an appropriate response message based on a judgment, A means for providing the generated response message by speech synthesis, A means of transmitting commands when immediate response by a human operator is required, A system that includes this.

2. The system according to claim 1, further comprising means for suggesting improvements to the route and operation of a machine in autonomous motion based on the classification and severity of the claims.

3. The system according to claim 1, further comprising means for monitoring the response results, generating a warning if the issue is not resolved, and notifying a human operator.

Images