system

Abstract

We provide the system. [Solution] A means for converting acquired audio data into text in real time, An artificial intelligence means that analyzes the converted text and generates an appropriate response, A means for displaying the generated response on the call responder's display device, 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, including 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] In modern society, call handling by phone is required by many companies and organizations. In particular, there is a problem that it is difficult for call handlers to respond quickly and accurately when complex questions or specialized knowledge are required. Furthermore, when a call handler makes a call in parallel with other tasks, the quality of the response may decline. In such a situation, there is a need for a system that converts the voice of the call partner into text in real time and presents an appropriate response.

Means for Solving the Problems

[0005] To solve this problem, the present invention provides the following means. First, it includes means for converting acquired voice data into text in real time. Next, it includes artificial intelligence means for analyzing the converted text and generating an appropriate response. This artificial intelligence has been pre-trained with industry-specific and department-specific knowledge and can handle specialized questions. Furthermore, it includes means for displaying the generated response on a display device for the call operator. The response displayed on the display device can be reviewed by the call operator and adjusted as needed, enabling flexible and efficient responses. This provides a system that reduces the burden on call operators and realizes high-quality call handling.

[0006] "Voice data" refers to acoustic signals acquired through telephones or other devices.

[0007] "Means of converting to text" refers to hardware or software functions that convert audio data into text information in real time.

[0008] "Artificial intelligence tools" refer to machine learning models and algorithms that analyze input text and generate appropriate responses based on its context and content.

[0009] A "display device" refers to a screen or monitor used to visually display the generated text response.

[0010] "Industry-specific knowledge" refers to specialized information and know-how related to a particular industry.

[0011] "Departmental knowledge" refers to specialized information and know-how related to the operations of a particular department.

[0012] "Analysis" refers to the process of understanding the meaning and context of input text.

[0013] "Response" refers to the answer or reply generated in response to an input question or request.

[0014] "Adjustable" means that the generated response can be changed or modified by the call responder at their own discretion.

[0015] "Call responder" refers to a person who is responsible for communicating with the other party via telephone.

Brief Explanation of Drawings

[0016] [Figure 1] It is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] It is a conceptual diagram showing an example of the main functions of a data processing device and a smart device according to the first embodiment. [Figure 3] It is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] It 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] It is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] It 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] It is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] It 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] It shows an emotion map to which a plurality of emotions are mapped. [Figure 10] It shows an emotion map to which a plurality of emotions are mapped. [Figure 11] It is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Example 2 when an emotion engine is combined. [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.

Mode for Carrying Out the Invention

[0017] 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.

[0018] First, the terms used in the following description will be explained.

[0019] In the following embodiments, a processor with a reference number (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be one 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.

[0020] In the following embodiments, a RAM (Random Access Memory) with a reference number is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0021] In the following embodiments, a storage with a reference number is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include a flash memory (SSD (Solid State Drive)), a magnetic disk (e.g., a hard disk), or a magnetic tape, etc.

[0022] 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).

[0023] 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."

[0024] [First Embodiment]

[0025] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0026] 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.

[0027] 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).

[0028] 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.

[0029] 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.

[0030] 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.

[0031] 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.

[0032] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0033] 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.

[0034] 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.

[0035] 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.

[0036] 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".

[0037] This invention provides a system that improves the efficiency and accuracy of telephone communication. This system realizes a technology that converts telephone audio into text in real time, inputs that text data into a generative artificial intelligence (AI) to generate an appropriate response, and displays it to the person handling the call.

[0038] System Configuration

[0039] This system mainly consists of the following three elements:

[0040] 1. Means for converting audio data into text (speech recognition engine)

[0041] 2. Response generation method using generative artificial intelligence (AI)

[0042] 3. Means for displaying response results (user interface)

[0043] Acquisition and conversion of audio data (server)

[0044] The server converts the audio data acquired during a call into text data in real time. The speech recognition engine can analyze the audio data and convert it into text information. In this process, the server uses a speech recognition API to output the audio data as text data.

[0045] Text analysis and response generation (server)

[0046] After text recognition processing, the server analyzes the recognized text data and inputs it into a generative AI. The generative AI generates appropriate responses to questions based on pre-learned industry and departmental knowledge. This AI utilizes state-of-the-art machine learning algorithms to provide highly accurate responses.

[0047] Display of the generated response (on the terminal)

[0048] The generated response is sent from the server to the user's terminal and displayed on the terminal's display. Because the response is displayed in real time for the call operator to review, the user can refer to it immediately.

[0049] Specific example

[0050] Here are some specific examples of its use:

[0051] Questions about registration

[0052] 1. Call initiated (User)

[0053] The user receives a phone call, and the caller asks, "How do I register for a new account?"

[0054] 2. Speech recognition (server)

[0055] The server converts the audio to text and generates the text, "Please tell me how to register for a new account."

[0056] 3. Response generation (server)

[0057] The server passes this text to the generative AI, which then generates a response saying, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0058] 4. Display of response (device)

[0059] The generated response is displayed on the user's monitor.

[0060] 5. Final interaction (user)

[0061] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "First, please visit our website and click the 'Register' button in the upper right corner. Then, please fill in the required information."

[0062] In this way, this system enables call operators to provide quick and accurate responses. By referring to responses generated by generative AI, call operators can immediately utilize their specialized knowledge, thereby improving the quality of their responses.

[0063] The following describes the processing flow.

[0064] Step 1: Acquiring audio data (user)

[0065] The user initiates communication with the other party via telephone. Voice data is acquired in real time through the microphone.

[0066] Step 2: Transferring audio data (to the device)

[0067] The terminal sends the acquired voice data to the server. The voice data is transferred in real time as data packets.

[0068] Step 3: Speech Recognition (Server)

[0069] The server receives the audio data and passes it to the speech recognition engine, which converts it into text data in real time. The speech recognition engine analyzes the audio data and generates corresponding text information.

[0070] Step 4: Text analysis and response generation (server)

[0071] The server analyzes the converted text data to understand its context and content. It then inputs this text into a generative artificial intelligence (AI) to generate an appropriate response. The generative AI utilizes pre-learned industry- and department-specific knowledge to produce highly accurate responses.

[0072] Step 5: Generating the response (server)

[0073] The generated response is reviewed and sent to the caller. The response is sent from the server to the terminal as text data.

[0074] Step 6: Displaying the response (on the device)

[0075] The terminal displays the received response on the caller's monitor. The display device shows the response in a highly visible and easy-to-understand format.

[0076] Step 7: Confirmation of response and response (user)

[0077] The user (call operator) reviews the displayed response. If necessary, the user adjusts the response and provides an appropriate final response to the other party. This allows the call operator to provide flexible and accurate responses while referring to the responses generated by the AI.

[0078] (Example 1)

[0079] 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."

[0080] Traditional call answering systems struggled to convert call audio into text in real time and generate appropriate responses. Furthermore, they lacked systems that enabled callers to instantly utilize their specialized knowledge to provide quick and accurate responses. As a result, the efficiency and accuracy of responses decreased, making it difficult to improve customer satisfaction.

[0081] 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.

[0082] In this invention, the server includes means for converting acquired audio data into text in real time, means for analyzing the converted text and inputting it as a prompt sentence into a generative AI model that has been pre-trained with industry-specific and department-specific knowledge, and means for displaying the response generated by the generative AI model on the call operator's display device. This makes it possible to instantly transcribe call audio into text and generate and display an appropriate response.

[0083] "Voice data" refers to digital signals obtained from voice, and is data used to record the content of a conversation in real time.

[0084] "Converting to text" refers to the process of analyzing acquired audio data and converting the linguistic information into string-based data.

[0085] A "generative AI model" is an artificial intelligence system that uses deep learning technology to dynamically generate appropriate responses to newly input text based on previously learned knowledge.

[0086] A "prompt sentence" is the initial text data input to a generative AI model, and it is the sentence that forms the basis for generating a response.

[0087] A "call operator" refers to a person who communicates with customers using a call system.

[0088] A "display device" is hardware used to visually display generated responses or other information, and includes monitors, displays, and screens.

[0089] "Real-time" refers to a state where processing and responses occur almost instantly, with virtually no delay or time lag.

[0090] "User interface means" refers to interactive screens or components that allow the call responder to review and adjust the generated response.

[0091] "Industry-specific and department-specific knowledge" refers to specialized information and knowledge related to a particular industry or department that has been learned in advance, forming the foundation for accurate and appropriate responses.

[0092] This invention provides a system that improves efficiency and accuracy in telephone communication. This system realizes a technology that converts telephone audio into text in real time, inputs that text data into a generating AI model to generate an appropriate response, and displays it to the person handling the call.

[0093] Acquisition and conversion of audio data

[0094] The server converts audio data acquired during a call into text data in real time. This process uses a speech recognition engine such as Google® Cloud Speech-to-Text API to analyze the audio data and convert the linguistic information into text data. For example, if a user says, "Please tell me how to register for a new account," the server converts this audio data into the text, "Please tell me how to register for a new account."

[0095] Text analysis and response generation

[0096] The server analyzes the acquired text data, identifies key keywords and context, and then inputs it into a generative AI model. Examples of generative AI models used here include OpenAI® GPT-3® and ChatGPT®. This model utilizes pre-trained industry-specific and department-specific knowledge to generate appropriate responses based on the text data provided as prompts. For example, in response to the prompt "How do I register for a new account?", it generates a response such as "To register for a new account, first visit our website and click the 'Register' button in the upper right corner."

[0097] Display of the generated response

[0098] The generated response is sent from the server to the user's terminal. The terminal has a web browser or dedicated application installed, which visually displays the received response on a display device. The display format and layout follow a pre-configured template, allowing the call operator to quickly review the response. The displayed response can also be adjusted via the user interface as needed.

[0099] Specific example

[0100] As a concrete example of its use, consider the following scenario:

[0101] 1. Call initiated (User)

[0102] The user receives a phone call and is asked, "How do I register for a new account?"

[0103] 2. Speech recognition (server)

[0104] The server converts the audio to text and generates the text, "Please tell me how to register for a new account."

[0105] 3. Response generation (server)

[0106] The server passes this text to the AI ​​model, which then generates the response, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0107] 4. Display of response (device)

[0108] The generated response is displayed on the user's monitor.

[0109] 5. Final interaction (user)

[0110] The caller will review the displayed text and, if necessary, provide a final response such as, "First, please visit our website and click the 'Register' button in the upper right corner. Then, please fill in the required information."

[0111] In this way, this system enables call operators to provide quick and accurate responses. By referring to responses generated by the AI ​​model, call operators can immediately utilize their specialized knowledge, thereby improving the quality of their responses.

[0112] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0113] System program processing flow

[0114] Step 1:

[0115] Acquisition of audio data

[0116] The server acquires voice data from the call system in real time. When a user says, "Please tell me how to register a new account," this voice data is sent to the server from the VoIP system or telephone system.

[0117] Input: Voice data from the call system

[0118] Output: Audio data file

[0119] Specific operation: The server receives the audio data and temporarily stores it.

[0120] Step 2:

[0121] Converting audio data to text

[0122] The server converts the acquired audio data into text data in real time. This process uses a speech recognition engine (for example, the Google Cloud Speech-to-Text API). The audio data is analyzed, and the text "How do I register for a new account?" is generated.

[0123] Input: Audio data

[0124] Output: Text data

[0125] Specific operation: The server calls a speech recognition API and outputs the audio data as text.

[0126] Step 3:

[0127] Text data analysis

[0128] The server analyzes the converted text data to identify key keywords and context. Based on this analysis, it prepares the data to be input as prompts to the generative AI model.

[0129] Input: Text data

[0130] Output: Analyzed text data

[0131] Specific operation: The server uses a parsing algorithm to analyze the structure and meaning of the text and format it into an appropriate prompt message.

[0132] Step 4:

[0133] Response generation using a generative AI model

[0134] The server inputs the analyzed text data into a generative AI model. This generative AI model (for example, OpenAI GPT-3) uses its pre-trained knowledge to generate an appropriate response such as, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0135] Input: Analyzed text data

[0136] Output: Generated response

[0137] Specific operation: The server sends a prompt to the generated AI model and receives the response in text format.

[0138] Step 5:

[0139] Preparing to display the generated response

[0140] The server receives the generated response and prepares to send it directly to the display device without going through the user interface. Display formatting and layout are applied, and the response is formatted in a way that is suitable for the user to read.

[0141] Input: Generated response

[0142] Output: Formatted text data for display

[0143] Specific operation: The server appropriately formats the response text and formats it according to the user interface template.

[0144] Step 6:

[0145] Display of the generated response

[0146] The server sends the formatted response data to the user's terminal. The terminal receives this data and displays it visually on its display device. The user (the person answering the call) reviews the displayed response and makes adjustments as needed.

[0147] Input: Formatted response data

[0148] Output: Response displayed on the display device

[0149] Specific operation: The server sends the formatted response data to the terminal, which then displays it on its display device. The user confirms this and verbally communicates it to the person they are talking to.

[0150] Through the above processing steps, it becomes possible to transcribe call audio into text in real time, generate appropriate responses using a generation AI model, and provide callers with quick and accurate information.

[0151] (Application Example 1)

[0152] 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."

[0153] Providing quick and accurate responses is crucial in telephone communication, but conventional methods make it difficult for callers to process large amounts of information instantly. Furthermore, callers need to shift their gaze to view display devices, which reduces the efficiency of the interaction. Additionally, generating responses tailored to specific industries or departments quickly is not easy. The objective of this invention is to solve these problems and improve the quality and efficiency of telephone communication.

[0154] 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.

[0155] In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the converted text and generating an appropriate response, means for displaying the generated response on a display device, and means for the display device to be integrated into a wearable device and to provide information to the call responder in real time. This allows the call responder to check the appropriate response in real time without moving their eyes and to respond immediately.

[0156] "Voice data" refers to the audio signals acquired during a phone call.

[0157] "Real-time" refers to processing that occurs at approximately the same rate as real-world time.

[0158] "Text" refers to the written information obtained by analyzing audio data.

[0159] "Conversion" refers to the process of converting audio data into text information.

[0160] "Analysis" refers to the process of understanding text data and extracting its meaning.

[0161] "Artificial intelligence tools" refer to algorithms and models that generate appropriate responses based on text data.

[0162] "Response" refers to an appropriate answer to a question or request, generated by artificial intelligence.

[0163] A "display device" refers to a device used to visually display the generated response.

[0164] A "wearable device" refers to a device that a user wears or uses on their body. Examples include smart glasses and head-mounted displays.

[0165] A "call handler" refers to a person who handles inquiries from customers or users.

[0166] "Providing information" refers to displaying the data and responses needed by the call operator on a display device in real time.

[0167] This invention is a system that improves the efficiency and accuracy of telephone communication. This system is characterized by converting acquired voice data into text in real time, analyzing the converted text to generate an appropriate response, and displaying the generated response on a display device of a wearable device.

[0168] Hardware and software to be used

[0169] Speech-to-Text API: Google Cloud Speech-to-Text

[0170] Generative artificial intelligence model: OpenAI GPT-4 (registered trademark)

[0171] Display device: Wearable device (e.g., smart glasses)

[0172] System configuration and operation

[0173] 1. Acquisition and conversion of audio data

[0174] The server converts the audio data acquired during a call into text in real time using the Google Cloud Speech-to-Text API. This ensures that the audio content of the call is instantly output as text.

[0175] 2. Text analysis and response generation

[0176] The server analyzes the converted text data and inputs it into OpenAI GPT-4. GPT-4, a generative AI model, generates appropriate responses based on pre-trained industry and sector knowledge.

[0177] Example of a prompt:

[0178] Please generate the best response to the following inquiry:

[0179] "My item hasn't arrived. What should I do?"

[0180] 3. Displaying the generated response

[0181] The generated response is sent from the server to the display device of a wearable device (smart glasses) and displayed for the caller to view in real time. This allows the caller to instantly grasp the information necessary to respond without having to shift their gaze significantly.

[0182] Specific example

[0183] scenario

[0184] 1. Call initiated (User)

[0185] A customer contacts a customer support representative wearing a wearable device and asks, "My product hasn't arrived. What should I do?"

[0186] 2. Speech recognition (server)

[0187] The server converts the customer's inquiry into text and generates the message, "My item hasn't arrived. What should I do?"

[0188] 3. Response generation (server)

[0189] The server inputs this text into a generative AI and generates the response, "We apologize for the inconvenience. If you could provide your order number, we will check it immediately."

[0190] 4. Response display (smart glasses)

[0191] The generated response is displayed on the smart glasses' screen, and the customer support representative reviews it.

[0192] 5. Final response (person in charge)

[0193] The staff member checks the response and replies to the customer, "We apologize for the inconvenience. Could you please tell us your order number?"

[0194] In this way, callers can check appropriate responses in real time through a wearable device without shifting their gaze, improving work efficiency and accuracy.

[0195] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0196] Step 1:

[0197] Acquisition of audio data

[0198] The user's call audio is captured through the microphone of the wearable device. The wearable device then sends this audio data to the server.

[0199] Input: User's call audio data

[0200] Output: Audio data sent to the server

[0201] Step 2:

[0202] Converting audio data to text

[0203] The server converts the audio data in real time using the Google Cloud Speech-to-Text API. In this process, the audio data is parsed via the API and the corresponding text data is output.

[0204] Input: Audio data

[0205] Output: Text data

[0206] Step 3:

[0207] Text data analysis

[0208] The server retrieves the converted text data and performs analysis to generate an appropriate response. This analysis understands the content of the text data and forms a response generation prompt based on that content.

[0209] Input: Text data

[0210] Output: Response generation prompt

[0211] Specific prompt message:

[0212] Please generate the best response to the following inquiry:

[0213] "My item hasn't arrived. What should I do?"

[0214] Step 4:

[0215] Response generation

[0216] The generative artificial intelligence model (OpenAI GPT-4) generates appropriate responses based on the generated prompts. In the response generation process, the generative AI model utilizes its pre-learned knowledge to produce highly accurate responses.

[0217] Input: Response generation prompt

[0218] Output: Generated response

[0219] Specific response:

[0220] "We apologize for the inconvenience. If you could provide your order number, we will check it immediately."

[0221] Step 5:

[0222] Sending the generated response

[0223] The generated response is sent from the server to the display device of the wearable device.

[0224] Input: Generated response

[0225] Output: Response data is sent to the wearable device.

[0226] Step 6:

[0227] Displaying the response

[0228] The wearable device displays the received response data on its screen. The response is displayed in real time so that the caller can see the content of the response.

[0229] Input: Response data

[0230] Output: The response displayed on the screen

[0231] Step 7:

[0232] Final response

[0233] The call operator will respond to the user appropriately based on the displayed response.

[0234] Input: Response displayed on the wearable device's screen.

[0235] Output: Final response to the user

[0236] Through this series of processing steps, users can respond to calls quickly and accurately.

[0237] 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.

[0238] This invention provides a system that improves efficiency and accuracy in telephone communication, and in particular, by combining it with a function that recognizes the user's emotions, it enables more appropriate responses. This system converts call audio into text in real time, inputs that text into a generative artificial intelligence (AI) to generate an appropriate response, and displays it to the caller. Furthermore, it incorporates an emotion engine that recognizes the user's emotions, and realizes technology that generates responses according to the emotional state.

[0239] System Configuration

[0240] This system mainly consists of the following four elements:

[0241] 1. Means for converting audio data into text (speech recognition engine)

[0242] 2. Response generation method using generative artificial intelligence (AI)

[0243] 3. Means for displaying the generated response (user interface)

[0244] 4. Means for recognizing the user's emotional state using an emotion engine

[0245] Acquisition and conversion of audio data (user and server)

[0246] The user communicates with the other party via telephone. During the call, audio data is transmitted in real time from the device to the server. The server uses a speech recognition engine to convert the audio data into text data. This process is performed using a speech recognition API.

[0247] Sentiment analysis (server)

[0248] The server uses an emotion engine to analyze the user's emotional state from the voice data. The emotion engine analyzes acoustic features such as tone, pitch, and speed of the voice to recognize the user's emotional state (e.g., stress, joy, anger, etc.). This recognition result is used in the subsequent response generation process.

[0249] Text analysis and response generation (server)

[0250] The server analyzes the text data converted by the speech recognition engine and inputs it into a generative artificial intelligence (AI). Based on pre-learned industry-specific and department-specific knowledge, the generative AI generates an appropriate response considering the text data and sentiment analysis results. This process enables flexible responses that respond to the user's emotions.

[0251] Display of the generated response (on the terminal)

[0252] The generated response is sent from the server to the user's terminal and displayed on the display device. The display device presents the information in a highly visible format that makes it easy for the user to confirm the response. This allows the call operator to immediately refer to the generated response.

[0253] Specific example

[0254] Here are some specific examples of its use:

[0255] Questions to ask when a user is feeling stressed.

[0256] 1. Call initiated (User)

[0257] The user receives a phone call, and the caller speaks in a strong tone, saying, "I am very dissatisfied with this service."

[0258] 2. Speech recognition and emotion recognition (server)

[0259] The server converts the speech to text and generates the text, "I am very dissatisfied with this service." At the same time, the emotion engine recognizes from the speech that the person is experiencing stress.

[0260] 3. Response generation (server)

[0261] The server passes this text and emotional state to the generative AI, which then generates a response such as, "I'm sorry. What part are you unhappy about?"

[0262] 4. Display of response (device)

[0263] The generated response is displayed on the user's monitor.

[0264] 5. Final interaction (user)

[0265] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "We apologize for the inconvenience. Specifically, what aspects are you dissatisfied with?"

[0266] Thus, by using this system, call operators can respond quickly and accurately while taking into account the user's emotional state. This is expected to improve the quality of service and increase user satisfaction.

[0267] The following describes the processing flow.

[0268] Step 1: Acquiring audio data (user)

[0269] The user initiates communication with the other party via telephone. Call audio data is captured in real time via the microphone.

[0270] Step 2: Transferring audio data (to the device)

[0271] The device sends the acquired voice data to the server. The voice data is transferred to the server in real time as data packets.

[0272] Step 3: Speech Recognition (Server)

[0273] The server receives the audio data and passes it to the speech recognition engine, which converts it into text data in real time. The speech recognition engine analyzes the audio data and converts it into text information.

[0274] Step 4: Sentiment Analysis (Server)

[0275] The server inputs the characteristics of the converted text and audio data into the emotion engine, which then analyzes the user's emotional state. The emotion engine recognizes the user's emotional state based on characteristics such as the tone, pitch, and speed of the voice.

[0276] Step 5: Text analysis and response generation (server)

[0277] The server inputs text data obtained through speech recognition and emotional information analyzed by the emotion engine into a generative artificial intelligence (AI) to generate an appropriate response. The generative AI utilizes industry-specific and departmental knowledge to derive the most appropriate response.

[0278] Step 6: Generating the response (server)

[0279] The generated responses are stored on the server as text data and adjusted as needed to match the user's emotional state.

[0280] Step 7: Send and view your reply (on your device)

[0281] The server sends the generated response to the terminal, which then displays it in real time on a display device connected to the terminal. The display device provides an intuitive and highly visible interface.

[0282] Step 8: Confirming the response and responding (to the user)

[0283] The user (call responder) checks the displayed response and makes minor corrections to the response content as necessary. By conveying the final response to the other party, an appropriate response according to the user's emotional state is realized.

[0284] As a specific example, when the user is feeling stressed and says, "I am very dissatisfied with this service," the server converts the voice into text, the emotion engine recognizes the stressed state, and the generative AI generates a response such as "We are very sorry for the inconvenience. Specifically, what points are you dissatisfied with?" and displays it to the user. After that, the user finally adjusts and responds with "Specifically, what points do you feel dissatisfied with?" In this way, this system realizes a more advanced call response by combining emotion recognition.

[0285] (Example 2)

[0286] Next, Example 2 will be described. In the following description, the data processing device 12 is referred to as the "server," and the smart device 14 is referred to as the "terminal."

[0287] In the conventional call response system, it was difficult to quickly provide a response considering the emotional state of the user during the call. As a result, communication between the call responder and the user did not proceed smoothly, and there was a possibility that dissatisfaction and stress would increase. Also, there was a problem that it was difficult to utilize industry - specific and department - specific expertise to generate appropriate responses, and the quality of responses was not consistent.

[0288] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following respective means.

[0289] In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the emotional state of the converted text and voice and generating an appropriate response, and means for displaying the generated response on a display device of the call responder. This enables the provision of appropriate responses that reflect the emotional state of the user during a call in real time, thereby improving the quality of call handling and increasing user satisfaction.

[0290] "Acquired audio data" refers to audio information collected through the user's speech during a call.

[0291] "Real-time" refers to processing or conversion happening almost instantly with minimal delay.

[0292] "Means of converting to text" refers to technical devices or software for converting audio data into textual information.

[0293] "Converted text" refers to the textual information extracted from audio data.

[0294] "Voice emotional state" refers to the emotional characteristics and state of the user analyzed from their voice, such as stress, joy, or anger.

[0295] "Analyzing" refers to the process of examining data in detail to understand its meaning and state.

[0296] "An artificial intelligence method for generating appropriate responses" refers to artificial intelligence technology that creates the optimal response based on the user's utterance and emotional state.

[0297] "Generated response" refers to a response text to a user's utterance that is created by artificial intelligence.

[0298] A "call answerer" refers to a person whose role is to answer phone calls and conversations from users.

[0299] The "display device" refers to a device for visually presenting the generated response and other information, such as a monitor or a tablet.

[0300] The "means for displaying" refers to a technical device or software that outputs the generated response to a display device so that the call responder can view it.

[0301] The "industry - and - department - specific knowledge" refers to specialized information and knowledge related to a specific industry or department.

[0302] "Pre - learned" refers to the state in which artificial intelligence has previously acquired and understood the data and information necessary for generating responses.

[0303] The "adjustable means" refers to a technical device or software that allows a call responder to modify or optimize the generated response.

[0304] Embodiments for Implementing the Invention

[0305] The present invention is a system that improves the efficiency and accuracy in call handling, and in particular, by combining the function of recognizing the user's emotions, enables more appropriate responses. This system is composed of the following four main elements:

[0306] 1. Means for real - time conversion of the acquired voice data into text

[0307] 2. Artificial intelligence means for analyzing the emotional states of the converted text and voice and generating appropriate responses

[0308] 3. Means for displaying the generated response on the display device of the call responder

[0309] 4. Means for allowing the call responder to adjust the displayed response

[0310] This system uses a voice recognition engine, an emotion engine, a generation AI model, and a display device.

[0311] Acquisition and conversion of audio data (user and server)

[0312] Users communicate with others via telephone. Audio data from the call is transmitted in real time to a server by the device (e.g., a smartphone or telephone). The server converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text or IBM Watson® Speech to Text). This process is performed using a speech recognition API.

[0313] Sentiment analysis (server)

[0314] The server uses an emotion engine (e.g., Affectiva or Microsoft® Azure® Emotion API) to analyze the user's emotional state from the voice data. The emotion engine analyzes acoustic features such as tone, pitch, and speed of the voice to recognize the user's emotional state (e.g., stress, joy, anger, etc.). This recognition result is used in the subsequent response generation process.

[0315] Text data analysis and response generation (server)

[0316] The server combines the converted text data with the sentiment analysis results and inputs them as prompts to a generative AI model (e.g., OpenAI GPT-4). For example, it might generate a prompt in the format of, "The user is feeling stressed. Please provide an appropriate response to 'I am very dissatisfied with this service.'" The generative AI model then generates an appropriate response based on this prompt.

[0317] Display and adjust the generated response (on the terminal)

[0318] The generated response is sent from the server to the user's device (e.g., a computer or tablet) and displayed on the device. The call operator can then review the displayed response and make adjustments as needed. For example, if the generated response is "I'm sorry, what exactly are you unhappy about?", they can adjust it to "I'm sorry for the inconvenience, but could you please elaborate on what specifically are you unhappy about?"

[0319] Specific example

[0320] Questions to ask when a user is feeling stressed.

[0321] 1. Call initiated (User)

[0322] The user receives a phone call, and the caller speaks in a strong tone, saying, "I am very dissatisfied with this service."

[0323] 2. Speech recognition and emotion recognition (server)

[0324] The server converts the speech to text and generates the text, "I am very dissatisfied with this service." At the same time, the emotion engine recognizes from the speech that the person is experiencing stress.

[0325] 3. Response generation (server)

[0326] The server passes this text and emotional state to the AI ​​model, which then generates a response such as, "I'm sorry. What part are you unhappy about?"

[0327] 4. Display of response (device)

[0328] The generated response is displayed on the user's monitor.

[0329] 5. Final interaction (user)

[0330] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "We apologize for the inconvenience. Specifically, what aspects are you dissatisfied with?"

[0331] This system allows call operators to respond quickly and accurately while considering the user's emotional state. This is expected to improve the quality of service and increase user satisfaction.

[0332] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0333] Step 1: Acquisition and transmission of audio data (user and device)

[0334] The user speaks over the phone. This audio data is captured by the device (e.g., a smartphone or telephone). The captured audio data is sent to the server in real time. The input is the user's voice, and the output is the audio data sent to the server.

[0335] Step 2: Convert audio data to text (server)

[0336] The server passes the received audio data to a speech recognition engine (for example, Google Cloud Speech-to-Text). The speech recognition engine converts the audio data into text data. At this stage, audio data is taken as input and text data is output. For example, the user's utterance is converted into the text "I am very dissatisfied with this service."

[0337] Step 3: Emotional state analysis (server)

[0338] The server passes the converted text data and raw audio data to an emotion engine (e.g., Microsoft Azure Emotion API). The emotion engine analyzes the user's emotional state from the audio data. At this stage, the input is the generated text data and audio data, and the output is the analysis result which includes the emotional state (e.g., the user is stressed).

[0339] Step 4: Generate prompt message (server)

[0340] The server generates prompts for the AI ​​model based on the acquired text data and emotional state. For example, it might generate a prompt such as, "The user is feeling stressed. Please provide an appropriate response to 'I am very dissatisfied with this service.'" The input is text data and emotional state, and the output is the generated prompt.

[0341] Step 5: Generating the response (server)

[0342] The server inputs a prompt into a generative AI model (e.g., OpenAI GPT-4). The generative AI model generates a response based on the prompt. The input is the prompt, and the output is an appropriate response to the user. For example, it might generate a response such as, "I'm sorry, but what part are you dissatisfied with?"

[0343] Step 6: Send the generated response (from server to terminal)

[0344] The generated response is sent from the server to the user's terminal. The input is the generated response, and the output is the data sent to the terminal.

[0345] Step 7: Display the generated response (on the terminal)

[0346] The generated response is displayed on a display device attached to the user's terminal (e.g., a monitor or tablet). The input is the response sent from the server, and the output is the text information displayed on the display device. Specifically, it might say, "We're sorry. What part are you dissatisfied with?"

[0347] Step 8: Final interaction (with the user)

[0348] The user reviews the displayed response and makes adjustments as needed. For example, the user might give a final response such as, "I'm sorry for the inconvenience. Specifically, what aspects are you dissatisfied with?" The input is the displayed response text, and the output is the user's final utterance.

[0349] (Application Example 2)

[0350] 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 device 14 will be referred to as the "terminal."

[0351] Traditional call answering systems often fail to consider the user's emotional state, leading to inadequate responses or inappropriate reactions. This can result in decreased user satisfaction and inconsistent service quality. Furthermore, it is difficult for call answerers to always provide the optimal response immediately, especially when dealing with emotionally charged users.

[0352] 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. In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the converted text and generating an appropriate response, emotion engine means for analyzing the user's emotional state, and means for displaying the response generated considering the emotion analysis results on the call operator's display device. This makes it possible to generate and provide an appropriate response to the call operator according to the user's emotional state.

[0353] "Voice data" refers to data that represents voice signals acquired during phone calls or voice communications in digital format.

[0354] "Real-time" refers to data processing and communication occurring almost instantly, resulting in minimal latency and instantaneous responsiveness.

[0355] "Text" refers to the string data that remains after audio data has been converted, and is expressed as written text in a form that humans can read.

[0356] "Artificial intelligence tools" are means of generating appropriate responses from input information using learning algorithms and data analysis techniques.

[0357] An "emotional engine" is a means of recognizing a user's emotional state by analyzing acoustic characteristics such as tone, pitch, and speed of speech.

[0358] A "call operator" refers to a person whose role is to respond to the other party through a phone call, and who uses the provided system or data to handle the call.

[0359] A "display device" is a device used to visually display generated responses or other information, and includes monitors and screens.

[0360] An embodiment of the present invention is a system that recognizes a user's emotions and generates and displays an optimal response according to that state. This system is particularly effective in user support for content distribution services, and responds quickly and accurately to user problems and questions through calls and chats.

[0361] System components

[0362] The system uses the following hardware and software:

[0363] hardware

[0364] Smartphone: Used as a device operated by the user, it acquires voice data and displays responses.

[0365] Server: A computer system used for analyzing voice data and generating responses.

[0366] software

[0367] Speech recognition APIs (e.g., Google Cloud Speech-to-Text): Convert speech data into text.

[0368] Emotion recognition engine (e.g., IBM Watson Tone Analyzer): Analyzes the user's emotional state from text.

[0369] Generative AI (e.g., OpenAI GPT-4): Generates appropriate responses based on the user's emotional state.

[0370] User Interface (UI): Displays the response generated on the smartphone or device.

[0371] System operation

[0372] The server uses a speech recognition API to convert audio data sent from the smartphone into text in real time. The converted text is then analyzed by an emotion recognition engine to identify the user's emotional state (e.g., stress, joy, anger). This emotion analysis is then used by a generative AI to generate an appropriate response. The generated response is sent from the server to the smartphone and displayed on the device.

[0373] Specific example

[0374] For example, if a user says, "The loading times are so slow, it's frustrating," the following process will occur:

[0375] 1. Acquisition and conversion of audio data

[0376] The smartphone's microphone captures audio, which is then sent to the server. The server uses the Google Cloud Speech-to-Text API to convert the audio data into text.

[0377] 2. Emotion analysis

[0378] The converted text is input into IBM Watson Tone Analyzer to recognize the user's emotional state (in this case, "frustration").

[0379] 3. Response generation

[0380] Based on the emotional state, OpenAI GPT-4 generates an appropriate response. For example, it might generate a response like, "We apologize for the inconvenience regarding loading times. We will address this immediately."

[0381] 4. Display the response

[0382] The generated response is sent from the server to the smartphone and displayed on the user interface.

[0383] Example of a prompt

[0384] User statement: "The loading times are so slow, it's frustrating."

[0385] User emotion: "Frustration"

[0386] Generate a supportive and empathetic response.

[0387] By using this system, it becomes possible to respond appropriately while taking user emotions into consideration, and it is expected that this will improve user satisfaction with content distribution services.

[0388] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0389] Program processing steps

[0390] Step 1:

[0391] The user uses their smartphone's microphone to input voice data. The smartphone acquires this voice data and sends it to the server in real time.

[0392] Input: User's voice

[0393] Output: Audio data sent to the server

[0394] Step 2:

[0395] The server inputs the received audio data into a speech recognition API and converts it into text data. For example, Google Cloud Speech-to-Text is used as the speech recognition API.

[0396] Input: Audio data sent to the server

[0397] Output: Text data

[0398] Step 3:

[0399] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotional state. IBM Watson Tone Analyzer is used as the emotion recognition engine.

[0400] Input: Text data

[0401] Output: User's emotional state (e.g., "frustration")

[0402] Step 4:

[0403] The server inputs the analyzed emotional state and text data into a generative AI to generate an appropriate response. OpenAI GPT-4 is used as the generative AI.

[0404] Input: Text data and user's emotional state

[0405] Output: Appropriate response

[0406] Step 5:

[0407] The server sends the generated response to the smartphone. The smartphone's user interface displays the received response.

[0408] Input: Generated response

[0409] Output: Response displayed on the smartphone's display screen

[0410] Specific operation of the process

[0411] Specific actions for Step 1:

[0412] When a user says to their smartphone, "The loading times are so slow, it's frustrating," the smartphone's microphone captures this audio and sends it to the server.

[0413] Specific actions in Step 2:

[0414] The server sends audio data to the Google Cloud Speech-to-Text API and receives this audio data as text. For example, the speech recognition API outputs the text "The loading time is so slow, it's frustrating."

[0415] Specific actions in Step 3:

[0416] The server sends the text message "The loading time is so slow, it's frustrating" to IBM Watson Tone Analyzer, and the analysis results in an emotional state of "frustration."

[0417] Specific actions in Step 4:

[0418] The server inputs the text data "The loading time is slow and frustrating" and the emotion state "frustrated" into OpenAI GPT-4 and generates a response such as "We apologize. We understand that you are experiencing inconvenience with the loading time. We will address this immediately."

[0419] Specific actions in Step 5:

[0420] The server sends the generated response to the smartphone. The smartphone's user interface then displays this response visually to the user.

[0421] In this way, a system can be realized that takes into account the user's emotions in real time and generates and displays appropriate responses.

[0422] 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.

[0423] 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 (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.

[0424] 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.

[0425] [Second Embodiment]

[0426] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0427] 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.

[0428] 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).

[0429] 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.

[0430] 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.

[0431] 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).

[0432] 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.

[0433] 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.

[0434] 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.

[0435] 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.

[0436] 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.

[0437] 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".

[0438] This invention provides a system that improves the efficiency and accuracy of telephone communication. This system realizes a technology that converts telephone audio into text in real time, inputs that text data into a generative artificial intelligence (AI) to generate an appropriate response, and displays it to the person handling the call.

[0439] System Configuration

[0440] This system mainly consists of the following three elements:

[0441] 1. Means for converting audio data into text (speech recognition engine)

[0442] 2. Response generation method using generative artificial intelligence (AI)

[0443] 3. Means for displaying response results (user interface)

[0444] Acquisition and conversion of audio data (server)

[0445] The server converts the audio data acquired during a call into text data in real time. The speech recognition engine can analyze the audio data and convert it into text information. In this process, the server uses a speech recognition API to output the audio data as text data.

[0446] Text analysis and response generation (server)

[0447] After text recognition processing, the server analyzes the recognized text data and inputs it into a generative AI. The generative AI generates appropriate responses to questions based on pre-learned industry and departmental knowledge. This AI utilizes state-of-the-art machine learning algorithms to provide highly accurate responses.

[0448] Display of the generated response (on the terminal)

[0449] The generated response is sent from the server to the user's terminal and displayed on the terminal's display. Because the response is displayed in real time for the call operator to review, the user can refer to it immediately.

[0450] Specific example

[0451] Here are some specific examples of its use:

[0452] Questions about registration

[0453] 1. Call initiated (User)

[0454] The user receives a phone call, and the caller asks, "How do I register for a new account?"

[0455] 2. Speech recognition (server)

[0456] The server converts the audio to text and generates the text, "Please tell me how to register for a new account."

[0457] 3. Response generation (server)

[0458] The server passes this text to the generative AI, which then generates a response saying, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0459] 4. Display of response (device)

[0460] The generated response is displayed on the user's monitor.

[0461] 5. Final interaction (user)

[0462] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "First, please visit our website and click the 'Register' button in the upper right corner. Then, please fill in the required information."

[0463] In this way, this system enables call operators to provide quick and accurate responses. By referring to responses generated by generative AI, call operators can immediately utilize their specialized knowledge, thereby improving the quality of their responses.

[0464] The following describes the processing flow.

[0465] Step 1: Acquiring audio data (user)

[0466] The user initiates communication with the other party via telephone. Voice data is acquired in real time through the microphone.

[0467] Step 2: Transferring audio data (to the device)

[0468] The terminal sends the acquired voice data to the server. The voice data is transferred in real time as data packets.

[0469] Step 3: Speech Recognition (Server)

[0470] The server receives the audio data and passes it to the speech recognition engine, which converts it into text data in real time. The speech recognition engine analyzes the audio data and generates corresponding text information.

[0471] Step 4: Text analysis and response generation (server)

[0472] The server analyzes the converted text data to understand its context and content. It then inputs this text into a generative artificial intelligence (AI) to generate an appropriate response. The generative AI utilizes pre-learned industry- and department-specific knowledge to produce highly accurate responses.

[0473] Step 5: Generating the response (server)

[0474] The generated response is reviewed and sent to the caller. The response is sent from the server to the terminal as text data.

[0475] Step 6: Displaying the response (on the device)

[0476] The terminal displays the received response on the caller's monitor. The display device shows the response in a highly visible and easy-to-understand format.

[0477] Step 7: Confirmation of response and response (user)

[0478] The user (call operator) reviews the displayed response. If necessary, the user adjusts the response and provides an appropriate final response to the other party. This allows the call operator to provide flexible and accurate responses while referring to the responses generated by the AI.

[0479] (Example 1)

[0480] 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".

[0481] Traditional call answering systems struggled to convert call audio into text in real time and generate appropriate responses. Furthermore, they lacked systems that enabled callers to instantly utilize their specialized knowledge to provide quick and accurate responses. As a result, the efficiency and accuracy of responses decreased, making it difficult to improve customer satisfaction.

[0482] 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.

[0483] In this invention, the server includes means for converting acquired audio data into text in real time, means for analyzing the converted text and inputting it as a prompt sentence into a generative AI model that has been pre-trained with industry-specific and department-specific knowledge, and means for displaying the response generated by the generative AI model on the call operator's display device. This makes it possible to instantly transcribe call audio into text and generate and display an appropriate response.

[0484] "Voice data" refers to digital signals obtained from voice, and is data used to record the content of a conversation in real time.

[0485] "Converting to text" refers to the process of analyzing acquired audio data and converting the linguistic information into string-based data.

[0486] A "generative AI model" is an artificial intelligence system that uses deep learning technology to dynamically generate appropriate responses to newly input text based on previously learned knowledge.

[0487] A "prompt sentence" is the initial text data input to a generative AI model, and it is the sentence that forms the basis for generating a response.

[0488] A "call operator" refers to a person who communicates with customers using a call system.

[0489] A "display device" is hardware used to visually display generated responses or other information, and includes monitors, displays, and screens.

[0490] "Real-time" refers to a state where processing and responses occur almost instantly, with virtually no delay or time lag.

[0491] "User interface means" refers to interactive screens or components that allow the call responder to review and adjust the generated response.

[0492] "Industry-specific and department-specific knowledge" refers to specialized information and knowledge related to a particular industry or department that has been learned in advance, forming the foundation for accurate and appropriate responses.

[0493] This invention provides a system that improves efficiency and accuracy in telephone communication. This system realizes a technology that converts telephone audio into text in real time, inputs that text data into a generating AI model to generate an appropriate response, and displays it to the person handling the call.

[0494] Acquisition and conversion of audio data

[0495] The server converts audio data acquired during a call into text data in real time. This process uses a speech recognition engine such as the Google Cloud Speech-to-Text API to analyze the audio data and convert the linguistic information into text data. For example, if a user says, "Please tell me how to register for a new account," the server converts this audio data into the text "Please tell me how to register for a new account."

[0496] Text analysis and response generation

[0497] The server analyzes the acquired text data, identifies key keywords and context, and then inputs it into a generative AI model. Examples of generative AI models used here include OpenAI GPT-3 and ChatGPT. This model leverages pre-trained industry- and department-specific knowledge to generate appropriate responses based on the text data provided as prompts. For example, in response to the prompt "How do I register for a new account?", it would generate a response such as "To register for a new account, first visit our website and click the 'Register' button in the upper right corner."

[0498] Display of the generated response

[0499] The generated response is sent from the server to the user's terminal. The terminal has a web browser or dedicated application installed, which visually displays the received response on a display device. The display format and layout follow a pre-configured template, allowing the call operator to quickly review the response. The displayed response can also be adjusted via the user interface as needed.

[0500] Specific example

[0501] As a concrete example of its use, consider the following scenario:

[0502] 1. Call initiated (User)

[0503] The user receives a phone call and is asked, "How do I register for a new account?"

[0504] 2. Speech recognition (server)

[0505] The server converts the audio to text and generates the text, "Please tell me how to register for a new account."

[0506] 3. Response generation (server)

[0507] The server passes this text to the AI ​​model, which then generates the response, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0508] 4. Display of response (device)

[0509] The generated response is displayed on the user's monitor.

[0510] 5. Final interaction (user)

[0511] The caller will review the displayed text and, if necessary, provide a final response such as, "First, please visit our website and click the 'Register' button in the upper right corner. Then, please fill in the required information."

[0512] In this way, this system enables call operators to provide quick and accurate responses. By referring to responses generated by the AI ​​model, call operators can immediately utilize their specialized knowledge, thereby improving the quality of their responses.

[0513] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0514] System program processing flow

[0515] Step 1:

[0516] Acquisition of audio data

[0517] The server acquires voice data from the call system in real time. When a user says, "Please tell me how to register a new account," this voice data is sent to the server from the VoIP system or telephone system.

[0518] Input: Voice data from the call system

[0519] Output: Audio data file

[0520] Specific operation: The server receives the audio data and temporarily stores it.

[0521] Step 2:

[0522] Converting audio data to text

[0523] The server converts the acquired audio data into text data in real time. This process uses a speech recognition engine (for example, the Google Cloud Speech-to-Text API). The audio data is analyzed, and the text "How do I register for a new account?" is generated.

[0524] Input: Audio data

[0525] Output: Text data

[0526] Specific operation: The server calls a speech recognition API and outputs the audio data as text.

[0527] Step 3:

[0528] Text data analysis

[0529] The server analyzes the converted text data to identify key keywords and context. Based on this analysis, it prepares the data to be input as prompts to the generative AI model.

[0530] Input: Text data

[0531] Output: Analyzed text data

[0532] Specific operation: The server uses a parsing algorithm to analyze the structure and meaning of the text and format it into an appropriate prompt message.

[0533] Step 4:

[0534] Response generation using a generative AI model

[0535] The server inputs the analyzed text data into a generative AI model. This generative AI model (for example, OpenAI GPT-3) uses its pre-trained knowledge to generate an appropriate response such as, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0536] Input: Analyzed text data

[0537] Output: Generated response

[0538] Specific operation: The server sends a prompt to the generated AI model and receives the response in text format.

[0539] Step 5:

[0540] Preparing to display the generated response

[0541] The server receives the generated response and prepares to send it directly to the display device without going through the user interface. Display formatting and layout are applied, and the response is formatted in a way that is suitable for the user to read.

[0542] Input: Generated response

[0543] Output: Formatted text data for display

[0544] Specific operation: The server appropriately formats the response text and formats it according to the user interface template.

[0545] Step 6:

[0546] Display of the generated response

[0547] The server sends the formatted response data to the user's terminal. The terminal receives this data and displays it visually on its display device. The user (the person answering the call) reviews the displayed response and makes adjustments as needed.

[0548] Input: Formatted response data

[0549] Output: Response displayed on the display device

[0550] Specific operation: The server sends the formatted response data to the terminal, which then displays it on its display device. The user confirms this and verbally communicates it to the person they are talking to.

[0551] Through the above processing steps, it becomes possible to transcribe call audio into text in real time, generate appropriate responses using a generation AI model, and provide callers with quick and accurate information.

[0552] (Application Example 1)

[0553] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0554] Providing quick and accurate responses is crucial in telephone communication, but conventional methods make it difficult for callers to process large amounts of information instantly. Furthermore, callers need to shift their gaze to view display devices, which reduces the efficiency of the interaction. Additionally, generating responses tailored to specific industries or departments quickly is not easy. The objective of this invention is to solve these problems and improve the quality and efficiency of telephone communication.

[0555] 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.

[0556] In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the converted text and generating an appropriate response, means for displaying the generated response on a display device, and means for the display device to be integrated into a wearable device and to provide information to the call responder in real time. This allows the call responder to check the appropriate response in real time without moving their eyes and to respond immediately.

[0557] "Voice data" refers to the audio signals acquired during a phone call.

[0558] "Real-time" refers to processing that occurs at approximately the same rate as real-world time.

[0559] "Text" refers to the written information obtained by analyzing audio data.

[0560] "Conversion" refers to the process of converting audio data into text information.

[0561] "Analysis" refers to the process of understanding text data and extracting its meaning.

[0562] "Artificial intelligence tools" refer to algorithms and models that generate appropriate responses based on text data.

[0563] "Response" refers to an appropriate answer to a question or request, generated by artificial intelligence.

[0564] A "display device" refers to a device used to visually display the generated response.

[0565] A "wearable device" refers to a device that a user wears or uses on their body. Examples include smart glasses and head-mounted displays.

[0566] A "call handler" refers to a person who handles inquiries from customers or users.

[0567] "Providing information" refers to displaying the data and responses needed by the call operator on a display device in real time.

[0568] This invention is a system that improves the efficiency and accuracy of telephone communication. This system is characterized by converting acquired voice data into text in real time, analyzing the converted text to generate an appropriate response, and displaying the generated response on a display device of a wearable device.

[0569] Hardware and software to be used

[0570] Speech-to-Text API: Google Cloud Speech-to-Text

[0571] Generative artificial intelligence model: OpenAI GPT-4

[0572] Display device: Wearable device (e.g., smart glasses)

[0573] System configuration and operation

[0574] 1. Acquisition and conversion of audio data

[0575] The server converts the audio data acquired during a call into text in real time using the Google Cloud Speech-to-Text API. This ensures that the audio content of the call is instantly output as text.

[0576] 2. Text analysis and response generation

[0577] The server analyzes the converted text data and inputs it into OpenAI GPT-4. GPT-4, a generative AI model, generates appropriate responses based on pre-trained industry and sector knowledge.

[0578] Example of a prompt:

[0579] Please generate the best response to the following inquiry:

[0580] "My item hasn't arrived. What should I do?"

[0581] 3. Displaying the generated response

[0582] The generated response is sent from the server to the display device of a wearable device (smart glasses) and displayed for the caller to view in real time. This allows the caller to instantly grasp the information necessary to respond without having to shift their gaze significantly.

[0583] Specific example

[0584] scenario

[0585] 1. Call initiated (User)

[0586] A customer contacts a customer support representative wearing a wearable device and asks, "My product hasn't arrived. What should I do?"

[0587] 2. Speech recognition (server)

[0588] The server converts the customer's inquiry into text and generates the message, "My item hasn't arrived. What should I do?"

[0589] 3. Response generation (server)

[0590] The server inputs this text into a generative AI and generates the response, "We apologize for the inconvenience. If you could provide your order number, we will check it immediately."

[0591] 4. Response display (smart glasses)

[0592] The generated response is displayed on the smart glasses' screen, and the customer support representative reviews it.

[0593] 5. Final response (person in charge)

[0594] The staff member checks the response and replies to the customer, "We apologize for the inconvenience. Could you please tell us your order number?"

[0595] In this way, callers can check appropriate responses in real time through a wearable device without shifting their gaze, improving work efficiency and accuracy.

[0596] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0597] Step 1:

[0598] Acquisition of audio data

[0599] The user's call audio is captured through the microphone of the wearable device. The wearable device then sends this audio data to the server.

[0600] Input: User's call audio data

[0601] Output: Audio data sent to the server

[0602] Step 2:

[0603] Converting audio data to text

[0604] The server converts the audio data in real time using the Google Cloud Speech-to-Text API. In this process, the audio data is parsed via the API and the corresponding text data is output.

[0605] Input: Audio data

[0606] Output: Text data

[0607] Step 3:

[0608] Text data analysis

[0609] The server retrieves the converted text data and performs analysis to generate an appropriate response. This analysis understands the content of the text data and forms a response generation prompt based on that content.

[0610] Input: Text data

[0611] Output: Response generation prompt

[0612] Specific prompt message:

[0613] Please generate the best response to the following inquiry:

[0614] "My item hasn't arrived. What should I do?"

[0615] Step 4:

[0616] Response generation

[0617] The generative artificial intelligence model (OpenAI GPT-4) generates appropriate responses based on the generated prompts. In the response generation process, the generative AI model utilizes its pre-learned knowledge to produce highly accurate responses.

[0618] Input: Response generation prompt

[0619] Output: Generated response

[0620] Specific response:

[0621] "We apologize for the inconvenience. If you could provide your order number, we will check it immediately."

[0622] Step 5:

[0623] Sending the generated response

[0624] The generated response is sent from the server to the display device of the wearable device.

[0625] Input: Generated response

[0626] Output: Response data is sent to the wearable device.

[0627] Step 6:

[0628] Displaying the response

[0629] The wearable device displays the received response data on its screen. The response is displayed in real time so that the caller can see the content of the response.

[0630] Input: Response data

[0631] Output: The response displayed on the screen

[0632] Step 7:

[0633] Final response

[0634] The call operator will respond to the user appropriately based on the displayed response.

[0635] Input: Response displayed on the wearable device's screen.

[0636] Output: Final response to the user

[0637] Through this series of processing steps, users can respond to calls quickly and accurately.

[0638] 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.

[0639] This invention provides a system that improves efficiency and accuracy in telephone communication, and in particular, by combining it with a function that recognizes the user's emotions, it enables more appropriate responses. This system converts call audio into text in real time, inputs that text into a generative artificial intelligence (AI) to generate an appropriate response, and displays it to the caller. Furthermore, it incorporates an emotion engine that recognizes the user's emotions, and realizes technology that generates responses according to the emotional state.

[0640] System Configuration

[0641] This system mainly consists of the following four elements:

[0642] 1. Means for converting audio data into text (speech recognition engine)

[0643] 2. Response generation method using generative artificial intelligence (AI)

[0644] 3. Means for displaying the generated response (user interface)

[0645] 4. Means for recognizing the user's emotional state using an emotion engine

[0646] Acquisition and conversion of audio data (user and server)

[0647] The user communicates with the other party via telephone. During the call, audio data is transmitted in real time from the device to the server. The server uses a speech recognition engine to convert the audio data into text data. This process is performed using a speech recognition API.

[0648] Sentiment analysis (server)

[0649] The server uses an emotion engine to analyze the user's emotional state from the voice data. The emotion engine analyzes acoustic features such as tone, pitch, and speed of the voice to recognize the user's emotional state (e.g., stress, joy, anger, etc.). This recognition result is used in the subsequent response generation process.

[0650] Text analysis and response generation (server)

[0651] The server analyzes the text data converted by the speech recognition engine and inputs it into a generative artificial intelligence (AI). Based on pre-learned industry-specific and department-specific knowledge, the generative AI generates an appropriate response considering the text data and sentiment analysis results. This process enables flexible responses that respond to the user's emotions.

[0652] Display of the generated response (on the terminal)

[0653] The generated response is sent from the server to the user's terminal and displayed on the display device. The display device presents the information in a highly visible format that makes it easy for the user to confirm the response. This allows the call operator to immediately refer to the generated response.

[0654] Specific example

[0655] Here are some specific examples of its use:

[0656] Questions to ask when a user is feeling stressed.

[0657] 1. Call initiated (User)

[0658] The user receives a phone call, and the caller speaks in a strong tone, saying, "I am very dissatisfied with this service."

[0659] 2. Speech recognition and emotion recognition (server)

[0660] The server converts the speech to text and generates the text, "I am very dissatisfied with this service." At the same time, the emotion engine recognizes from the speech that the person is experiencing stress.

[0661] 3. Response generation (server)

[0662] The server passes this text and emotional state to the generative AI, which then generates a response such as, "I'm sorry. What part are you unhappy about?"

[0663] 4. Display of response (device)

[0664] The generated response is displayed on the user's monitor.

[0665] 5. Final interaction (user)

[0666] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "We apologize for the inconvenience. Specifically, what aspects are you dissatisfied with?"

[0667] Thus, by using this system, call operators can respond quickly and accurately while taking into account the user's emotional state. This is expected to improve the quality of service and increase user satisfaction.

[0668] The following describes the processing flow.

[0669] Step 1: Acquiring audio data (user)

[0670] The user initiates communication with the other party via telephone. Call audio data is captured in real time via the microphone.

[0671] Step 2: Transferring audio data (to the device)

[0672] The device sends the acquired voice data to the server. The voice data is transferred to the server in real time as data packets.

[0673] Step 3: Speech Recognition (Server)

[0674] The server receives the audio data and passes it to the speech recognition engine, which converts it into text data in real time. The speech recognition engine analyzes the audio data and converts it into text information.

[0675] Step 4: Sentiment Analysis (Server)

[0676] The server inputs the characteristics of the converted text and audio data into the emotion engine, which then analyzes the user's emotional state. The emotion engine recognizes the user's emotional state based on characteristics such as the tone, pitch, and speed of the voice.

[0677] Step 5: Text analysis and response generation (server)

[0678] The server inputs text data obtained through speech recognition and emotional information analyzed by the emotion engine into a generative artificial intelligence (AI) to generate an appropriate response. The generative AI utilizes industry-specific and departmental knowledge to derive the most appropriate response.

[0679] Step 6: Generating the response (server)

[0680] The generated responses are stored on the server as text data and adjusted as needed to match the user's emotional state.

[0681] Step 7: Send and view your reply (on your device)

[0682] The server sends the generated response to the terminal, which then displays it in real time on a display device connected to the terminal. The display device provides an intuitive and highly visible interface.

[0683] Step 8: Confirming the response and responding (to the user)

[0684] The user (the person answering the call) reviews the displayed response and makes minor adjustments to the response content as needed. By communicating the final response to the other party, an appropriate response is achieved that is tailored to the user's emotional state.

[0685] For example, if a user is feeling stressed and says, "I'm very dissatisfied with this service," the server converts the speech to text, the emotion engine recognizes the user's stress level, and the generative AI generates a response such as, "We are very sorry for the inconvenience. Specifically, what aspects are you dissatisfied with?" and displays it to the user. The user then refines the response by asking, "Specifically, what aspects are you dissatisfied with?" In this way, the system achieves more sophisticated call handling by combining emotion recognition.

[0686] (Example 2)

[0687] 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".

[0688] Traditional call answering systems struggled to provide timely responses that took into account the emotional state of the user during a call. As a result, communication between the call answerer and the user was often cumbersome, potentially leading to increased frustration and stress. Furthermore, it was difficult to leverage industry-specific and departmental expertise to generate appropriate responses, resulting in inconsistent response quality.

[0689] 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.

[0690] In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the emotional state of the converted text and voice and generating an appropriate response, and means for displaying the generated response on a display device of the call responder. This enables the provision of appropriate responses that reflect the emotional state of the user during a call in real time, thereby improving the quality of call handling and increasing user satisfaction.

[0691] "Acquired audio data" refers to audio information collected through the user's speech during a call.

[0692] "Real-time" refers to processing or conversion happening almost instantly with minimal delay.

[0693] "Means of converting to text" refers to technical devices or software for converting audio data into textual information.

[0694] "Converted text" refers to the textual information extracted from audio data.

[0695] "Voice emotional state" refers to the emotional characteristics and state of the user analyzed from their voice, such as stress, joy, or anger.

[0696] "Analyzing" refers to the process of examining data in detail to understand its meaning and state.

[0697] "An artificial intelligence method for generating appropriate responses" refers to artificial intelligence technology that creates the optimal response based on the user's utterance and emotional state.

[0698] "Generated response" refers to a response text to a user's utterance that is created by artificial intelligence.

[0699] A "call answerer" refers to a person whose role is to answer phone calls and conversations from users.

[0700] A "display device" refers to a device used to visually show generated responses or other information, such as a monitor or tablet.

[0701] "Means of display" refers to a technical device or software that outputs the generated response to a display device so that the caller can confirm it.

[0702] "Industry-specific and departmental knowledge" refers to specialized information and insights related to a particular industry or department.

[0703] "Prior learning" refers to a state where artificial intelligence has already acquired and understood the data and information necessary to generate a response.

[0704] "Means of adjustment" refers to technical devices or software that allow the call responder to modify or optimize the generated response.

[0705] Modes for carrying out the invention

[0706] This invention is a system that improves the efficiency and accuracy of call handling, and in particular, by combining it with a function that recognizes the user's emotions, it enables more appropriate responses. This system consists of the following four main elements:

[0707] 1. A means of converting acquired audio data into text in real time.

[0708] 2. An artificial intelligence means for analyzing the emotional state of converted text and audio and generating appropriate responses.

[0709] 3. Means for displaying the generated response on the caller's display device.

[0710] 4. A means by which the caller can adjust the displayed response.

[0711] This system uses a speech recognition engine, an emotion engine, a generative AI model, and a display device.

[0712] Acquisition and conversion of audio data (user and server)

[0713] Users communicate with others via telephone. Audio data from the call is transmitted in real time to a server by the device (e.g., a smartphone or telephone). The server converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text or IBM Watson Speech to Text). This process is performed using speech recognition APIs.

[0714] Sentiment analysis (server)

[0715] The server uses an emotion engine (e.g., Affectiva or Microsoft Azure Emotion API) to analyze the user's emotional state from the voice data. The emotion engine analyzes acoustic features such as tone, pitch, and speed of the voice to recognize the user's emotional state (e.g., stress, joy, anger, etc.). This recognition result is then used in the subsequent response generation process.

[0716] Text data analysis and response generation (server)

[0717] The server combines the converted text data with the sentiment analysis results and inputs them as prompts to a generative AI model (e.g., OpenAI GPT-4). For example, it might generate a prompt in the format of, "The user is feeling stressed. Please provide an appropriate response to 'I am very dissatisfied with this service.'" The generative AI model then generates an appropriate response based on this prompt.

[0718] Display and adjust the generated response (on the terminal)

[0719] The generated response is sent from the server to the user's device (e.g., a computer or tablet) and displayed on the device. The call operator can then review the displayed response and make adjustments as needed. For example, if the generated response is "I'm sorry, what exactly are you unhappy about?", they can adjust it to "I'm sorry for the inconvenience, but could you please elaborate on what specifically are you unhappy about?"

[0720] Specific example

[0721] Questions to ask when a user is feeling stressed.

[0722] 1. Call initiated (User)

[0723] The user receives a phone call, and the caller speaks in a strong tone, saying, "I am very dissatisfied with this service."

[0724] 2. Speech recognition and emotion recognition (server)

[0725] The server converts the speech to text and generates the text, "I am very dissatisfied with this service." At the same time, the emotion engine recognizes from the speech that the person is experiencing stress.

[0726] 3. Response generation (server)

[0727] The server passes this text and emotional state to the AI ​​model, which then generates a response such as, "I'm sorry. What part are you unhappy about?"

[0728] 4. Display of response (device)

[0729] The generated response is displayed on the user's monitor.

[0730] 5. Final interaction (user)

[0731] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "We apologize for the inconvenience. Specifically, what aspects are you dissatisfied with?"

[0732] This system allows call operators to respond quickly and accurately while considering the user's emotional state. This is expected to improve the quality of service and increase user satisfaction.

[0733] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0734] Step 1: Acquisition and transmission of audio data (user and device)

[0735] The user speaks over the phone. This audio data is captured by the device (e.g., a smartphone or telephone). The captured audio data is sent to the server in real time. The input is the user's voice, and the output is the audio data sent to the server.

[0736] Step 2: Convert audio data to text (server)

[0737] The server passes the received audio data to a speech recognition engine (for example, Google Cloud Speech-to-Text). The speech recognition engine converts the audio data into text data. At this stage, audio data is taken as input and text data is output. For example, the user's utterance is converted into the text "I am very dissatisfied with this service."

[0738] Step 3: Emotional state analysis (server)

[0739] The server passes the converted text data and raw audio data to an emotion engine (e.g., Microsoft Azure Emotion API). The emotion engine analyzes the user's emotional state from the audio data. At this stage, the input is the generated text data and audio data, and the output is the analysis result which includes the emotional state (e.g., the user is stressed).

[0740] Step 4: Generate prompt message (server)

[0741] The server generates prompts for the AI ​​model based on the acquired text data and emotional state. For example, it might generate a prompt such as, "The user is feeling stressed. Please provide an appropriate response to 'I am very dissatisfied with this service.'" The input is text data and emotional state, and the output is the generated prompt.

[0742] Step 5: Generating the response (server)

[0743] The server inputs a prompt into a generative AI model (e.g., OpenAI GPT-4). The generative AI model generates a response based on the prompt. The input is the prompt, and the output is an appropriate response to the user. For example, it might generate a response such as, "I'm sorry, but what part are you dissatisfied with?"

[0744] Step 6: Send the generated response (from server to terminal)

[0745] The generated response is sent from the server to the user's terminal. The input is the generated response, and the output is the data sent to the terminal.

[0746] Step 7: Display the generated response (on the terminal)

[0747] The generated response is displayed on a display device attached to the user's terminal (e.g., a monitor or tablet). The input is the response sent from the server, and the output is the text information displayed on the display device. Specifically, it might say, "We're sorry. What part are you dissatisfied with?"

[0748] Step 8: Final interaction (with the user)

[0749] The user reviews the displayed response and makes adjustments as needed. For example, the user might give a final response such as, "I'm sorry for the inconvenience. Specifically, what aspects are you dissatisfied with?" The input is the displayed response text, and the output is the user's final utterance.

[0750] (Application Example 2)

[0751] 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."

[0752] Traditional call answering systems often fail to consider the user's emotional state, leading to inadequate responses or inappropriate reactions. This can result in decreased user satisfaction and inconsistent service quality. Furthermore, it is difficult for call answerers to always provide the optimal response immediately, especially when dealing with emotionally charged users.

[0753] 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. In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the converted text and generating an appropriate response, emotion engine means for analyzing the user's emotional state, and means for displaying the response generated considering the emotion analysis results on the call operator's display device. This makes it possible to generate and provide an appropriate response to the call operator according to the user's emotional state.

[0754] "Voice data" refers to data that represents voice signals acquired during phone calls or voice communications in digital format.

[0755] "Real-time" refers to data processing and communication occurring almost instantly, resulting in minimal latency and instantaneous responsiveness.

[0756] "Text" refers to the string data that remains after audio data has been converted, and is expressed as written text in a form that humans can read.

[0757] "Artificial intelligence tools" are means of generating appropriate responses from input information using learning algorithms and data analysis techniques.

[0758] An "emotional engine" is a means of recognizing a user's emotional state by analyzing acoustic characteristics such as tone, pitch, and speed of speech.

[0759] A "call operator" refers to a person whose role is to respond to the other party through a phone call, and who uses the provided system or data to handle the call.

[0760] A "display device" is a device used to visually display generated responses or other information, and includes monitors and screens.

[0761] An embodiment of the present invention is a system that recognizes a user's emotions and generates and displays an optimal response according to that state. This system is particularly effective in user support for content distribution services, and responds quickly and accurately to user problems and questions through calls and chats.

[0762] System components

[0763] The system uses the following hardware and software:

[0764] hardware

[0765] Smartphone: Used as a device operated by the user, it acquires voice data and displays responses.

[0766] Server: A computer system used for analyzing voice data and generating responses.

[0767] software

[0768] Speech recognition APIs (e.g., Google Cloud Speech-to-Text): Convert speech data into text.

[0769] Emotion recognition engine (e.g., IBM Watson Tone Analyzer): Analyzes the user's emotional state from text.

[0770] Generative AI (e.g., OpenAI GPT-4): Generates appropriate responses based on the user's emotional state.

[0771] User Interface (UI): Displays the response generated on the smartphone or device.

[0772] System operation

[0773] The server uses a speech recognition API to convert audio data sent from the smartphone into text in real time. The converted text is then analyzed by an emotion recognition engine to identify the user's emotional state (e.g., stress, joy, anger). This emotion analysis is then used by a generative AI to generate an appropriate response. The generated response is sent from the server to the smartphone and displayed on the device.

[0774] Specific example

[0775] For example, if a user says, "The loading times are so slow, it's frustrating," the following process will occur:

[0776] 1. Acquisition and conversion of audio data

[0777] The smartphone's microphone captures audio, which is then sent to the server. The server uses the Google Cloud Speech-to-Text API to convert the audio data into text.

[0778] 2. Emotion analysis

[0779] The converted text is input into IBM Watson Tone Analyzer to recognize the user's emotional state (in this case, "frustration").

[0780] 3. Response generation

[0781] Based on the emotional state, OpenAI GPT-4 generates an appropriate response. For example, it might generate a response like, "We apologize for the inconvenience regarding loading times. We will address this immediately."

[0782] 4. Display the response

[0783] The generated response is sent from the server to the smartphone and displayed on the user interface.

[0784] Example of a prompt

[0785] User statement: "The loading times are so slow, it's frustrating."

[0786] User emotion: "Frustration"

[0787] Generate a supportive and empathetic response.

[0788] By using this system, it becomes possible to respond appropriately while taking user emotions into consideration, and it is expected that this will improve user satisfaction with content distribution services.

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

[0790] Program processing steps

[0791] Step 1:

[0792] The user uses their smartphone's microphone to input voice data. The smartphone acquires this voice data and sends it to the server in real time.

[0793] Input: User's voice

[0794] Output: Audio data sent to the server

[0795] Step 2:

[0796] The server inputs the received audio data into a speech recognition API and converts it into text data. For example, Google Cloud Speech-to-Text is used as the speech recognition API.

[0797] Input: Audio data sent to the server

[0798] Output: Text data

[0799] Step 3:

[0800] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotional state. IBM Watson Tone Analyzer is used as the emotion recognition engine.

[0801] Input: Text data

[0802] Output: User's emotional state (e.g., "frustration")

[0803] Step 4:

[0804] The server inputs the analyzed emotional state and text data into a generative AI to generate an appropriate response. OpenAI GPT-4 is used as the generative AI.

[0805] Input: Text data and user's emotional state

[0806] Output: Appropriate response

[0807] Step 5:

[0808] The server sends the generated response to the smartphone. The smartphone's user interface displays the received response.

[0809] Input: Generated response

[0810] Output: Response displayed on the smartphone's display screen

[0811] Specific operation of the process

[0812] Specific actions for Step 1:

[0813] When a user says to their smartphone, "The loading times are so slow, it's frustrating," the smartphone's microphone captures this audio and sends it to the server.

[0814] Specific actions in Step 2:

[0815] The server sends audio data to the Google Cloud Speech-to-Text API and receives this audio data as text. For example, the speech recognition API outputs the text "The loading time is so slow, it's frustrating."

[0816] Specific actions in Step 3:

[0817] The server sends the text message "The loading time is so slow, it's frustrating" to IBM Watson Tone Analyzer, and the analysis results in an emotional state of "frustration."

[0818] Specific actions in Step 4:

[0819] The server inputs the text data "The loading time is slow and frustrating" and the emotion state "frustrated" into OpenAI GPT-4 and generates a response such as "We apologize. We understand that you are experiencing inconvenience with the loading time. We will address this immediately."

[0820] Specific actions in Step 5:

[0821] The server sends the generated response to the smartphone. The smartphone's user interface then displays this response visually to the user.

[0822] In this way, a system can be realized that takes into account the user's emotions in real time and generates and displays appropriate responses.

[0823] 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.

[0824] 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.

[0825] 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.

[0826] [Third Embodiment]

[0827] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0828] 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.

[0829] 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).

[0830] 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.

[0831] 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.

[0832] 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).

[0833] 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.

[0834] 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.

[0835] 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.

[0836] 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.

[0837] 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.

[0838] 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".

[0839] This invention provides a system that improves the efficiency and accuracy of telephone communication. This system realizes a technology that converts telephone audio into text in real time, inputs that text data into a generative artificial intelligence (AI) to generate an appropriate response, and displays it to the person handling the call.

[0840] System Configuration

[0841] This system mainly consists of the following three elements:

[0842] 1. Means for converting audio data into text (speech recognition engine)

[0843] 2. Response generation method using generative artificial intelligence (AI)

[0844] 3. Means for displaying response results (user interface)

[0845] Acquisition and conversion of audio data (server)

[0846] The server converts the audio data acquired during a call into text data in real time. The speech recognition engine can analyze the audio data and convert it into text information. In this process, the server uses a speech recognition API to output the audio data as text data.

[0847] Text analysis and response generation (server)

[0848] After text recognition processing, the server analyzes the recognized text data and inputs it into a generative AI. The generative AI generates appropriate responses to questions based on pre-learned industry and departmental knowledge. This AI utilizes state-of-the-art machine learning algorithms to provide highly accurate responses.

[0849] Display of the generated response (on the terminal)

[0850] The generated response is sent from the server to the user's terminal and displayed on the terminal's display. Because the response is displayed in real time for the call operator to review, the user can refer to it immediately.

[0851] Specific example

[0852] Here are some specific examples of its use:

[0853] Questions about registration

[0854] 1. Call initiated (User)

[0855] The user receives a phone call, and the caller asks, "How do I register for a new account?"

[0856] 2. Speech recognition (server)

[0857] The server converts the audio to text and generates the text, "Please tell me how to register for a new account."

[0858] 3. Response generation (server)

[0859] The server passes this text to the generative AI, which then generates a response saying, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0860] 4. Display of response (device)

[0861] The generated response is displayed on the user's monitor.

[0862] 5. Final interaction (user)

[0863] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "First, please visit our website and click the 'Register' button in the upper right corner. Then, please fill in the required information."

[0864] In this way, this system enables call operators to provide quick and accurate responses. By referring to responses generated by generative AI, call operators can immediately utilize their specialized knowledge, thereby improving the quality of their responses.

[0865] The following describes the processing flow.

[0866] Step 1: Acquiring audio data (user)

[0867] The user initiates communication with the other party via telephone. Voice data is acquired in real time through the microphone.

[0868] Step 2: Transferring audio data (to the device)

[0869] The terminal sends the acquired voice data to the server. The voice data is transferred in real time as data packets.

[0870] Step 3: Speech Recognition (Server)

[0871] The server receives the audio data and passes it to the speech recognition engine, which converts it into text data in real time. The speech recognition engine analyzes the audio data and generates corresponding text information.

[0872] Step 4: Text analysis and response generation (server)

[0873] The server analyzes the converted text data to understand its context and content. It then inputs this text into a generative artificial intelligence (AI) to generate an appropriate response. The generative AI utilizes pre-learned industry- and department-specific knowledge to produce highly accurate responses.

[0874] Step 5: Generating the response (server)

[0875] The generated response is reviewed and sent to the caller. The response is sent from the server to the terminal as text data.

[0876] Step 6: Displaying the response (on the device)

[0877] The terminal displays the received response on the caller's monitor. The display device shows the response in a highly visible and easy-to-understand format.

[0878] Step 7: Confirmation of response and response (user)

[0879] The user (call operator) reviews the displayed response. If necessary, the user adjusts the response and provides an appropriate final response to the other party. This allows the call operator to provide flexible and accurate responses while referring to the responses generated by the AI.

[0880] (Example 1)

[0881] 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."

[0882] Traditional call answering systems struggled to convert call audio into text in real time and generate appropriate responses. Furthermore, they lacked systems that enabled callers to instantly utilize their specialized knowledge to provide quick and accurate responses. As a result, the efficiency and accuracy of responses decreased, making it difficult to improve customer satisfaction.

[0883] 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.

[0884] In this invention, the server includes means for converting acquired audio data into text in real time, means for analyzing the converted text and inputting it as a prompt sentence into a generative AI model that has been pre-trained with industry-specific and department-specific knowledge, and means for displaying the response generated by the generative AI model on the call operator's display device. This makes it possible to instantly transcribe call audio into text and generate and display an appropriate response.

[0885] "Voice data" refers to digital signals obtained from voice, and is data used to record the content of a conversation in real time.

[0886] "Converting to text" refers to the process of analyzing acquired audio data and converting the linguistic information into string-based data.

[0887] A "generative AI model" is an artificial intelligence system that uses deep learning technology to dynamically generate appropriate responses to newly input text based on previously learned knowledge.

[0888] A "prompt sentence" is the initial text data input to a generative AI model, and it is the sentence that forms the basis for generating a response.

[0889] A "call operator" refers to a person who communicates with customers using a call system.

[0890] A "display device" is hardware used to visually display generated responses or other information, and includes monitors, displays, and screens.

[0891] "Real-time" refers to a state where processing and responses occur almost instantly, with virtually no delay or time lag.

[0892] "User interface means" refers to interactive screens or components that allow the call responder to review and adjust the generated response.

[0893] "Industry-specific and department-specific knowledge" refers to specialized information and knowledge related to a particular industry or department that has been learned in advance, forming the foundation for accurate and appropriate responses.

[0894] This invention provides a system that improves efficiency and accuracy in telephone communication. This system realizes a technology that converts telephone audio into text in real time, inputs that text data into a generating AI model to generate an appropriate response, and displays it to the person handling the call.

[0895] Acquisition and conversion of audio data

[0896] The server converts audio data acquired during a call into text data in real time. This process uses a speech recognition engine such as the Google Cloud Speech-to-Text API to analyze the audio data and convert the linguistic information into text data. For example, if a user says, "Please tell me how to register for a new account," the server converts this audio data into the text "Please tell me how to register for a new account."

[0897] Text analysis and response generation

[0898] The server analyzes the acquired text data, identifies key keywords and context, and then inputs it into a generative AI model. Examples of generative AI models used here include OpenAI GPT-3 and ChatGPT. This model leverages pre-trained industry- and department-specific knowledge to generate appropriate responses based on the text data provided as prompts. For example, in response to the prompt "How do I register for a new account?", it would generate a response such as "To register for a new account, first visit our website and click the 'Register' button in the upper right corner."

[0899] Display of the generated response

[0900] The generated response is sent from the server to the user's terminal. The terminal has a web browser or dedicated application installed, which visually displays the received response on a display device. The display format and layout follow a pre-configured template, allowing the call operator to quickly review the response. The displayed response can also be adjusted via the user interface as needed.

[0901] Specific example

[0902] As a concrete example of its use, consider the following scenario:

[0903] 1. Call initiated (User)

[0904] The user receives a phone call and is asked, "How do I register for a new account?"

[0905] 2. Speech recognition (server)

[0906] The server converts the audio to text and generates the text, "Please tell me how to register for a new account."

[0907] 3. Response generation (server)

[0908] The server passes this text to the AI ​​model, which then generates the response, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0909] 4. Display of response (device)

[0910] The generated response is displayed on the user's monitor.

[0911] 5. Final interaction (user)

[0912] The caller will review the displayed text and, if necessary, provide a final response such as, "First, please visit our website and click the 'Register' button in the upper right corner. Then, please fill in the required information."

[0913] In this way, this system enables call operators to provide quick and accurate responses. By referring to responses generated by the AI ​​model, call operators can immediately utilize their specialized knowledge, thereby improving the quality of their responses.

[0914] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0915] System program processing flow

[0916] Step 1:

[0917] Acquisition of audio data

[0918] The server acquires voice data from the call system in real time. When a user says, "Please tell me how to register a new account," this voice data is sent to the server from the VoIP system or telephone system.

[0919] Input: Voice data from the call system

[0920] Output: Audio data file

[0921] Specific operation: The server receives the audio data and temporarily stores it.

[0922] Step 2:

[0923] Converting audio data to text

[0924] The server converts the acquired audio data into text data in real time. This process uses a speech recognition engine (for example, the Google Cloud Speech-to-Text API). The audio data is analyzed, and the text "How do I register for a new account?" is generated.

[0925] Input: Audio data

[0926] Output: Text data

[0927] Specific operation: The server calls a speech recognition API and outputs the audio data as text.

[0928] Step 3:

[0929] Text data analysis

[0930] The server analyzes the converted text data to identify key keywords and context. Based on this analysis, it prepares the data to be input as prompts to the generative AI model.

[0931] Input: Text data

[0932] Output: Analyzed text data

[0933] Specific operation: The server uses a parsing algorithm to analyze the structure and meaning of the text and format it into an appropriate prompt message.

[0934] Step 4:

[0935] Response generation using a generative AI model

[0936] The server inputs the analyzed text data into a generative AI model. This generative AI model (for example, OpenAI GPT-3) uses its pre-trained knowledge to generate an appropriate response such as, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[0937] Input: Analyzed text data

[0938] Output: Generated response

[0939] Specific operation: The server sends a prompt to the generated AI model and receives the response in text format.

[0940] Step 5:

[0941] Preparing to display the generated response

[0942] The server receives the generated response and prepares to send it directly to the display device without going through the user interface. Display formatting and layout are applied, and the response is formatted in a way that is suitable for the user to read.

[0943] Input: Generated response

[0944] Output: Formatted text data for display

[0945] Specific operation: The server appropriately formats the response text and formats it according to the user interface template.

[0946] Step 6:

[0947] Display of the generated response

[0948] The server sends the formatted response data to the user's terminal. The terminal receives this data and displays it visually on its display device. The user (the person answering the call) reviews the displayed response and makes adjustments as needed.

[0949] Input: Formatted response data

[0950] Output: Response displayed on the display device

[0951] Specific operation: The server sends the formatted response data to the terminal, which then displays it on its display device. The user confirms this and verbally communicates it to the person they are talking to.

[0952] Through the above processing steps, it becomes possible to transcribe call audio into text in real time, generate appropriate responses using a generation AI model, and provide callers with quick and accurate information.

[0953] (Application Example 1)

[0954] 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."

[0955] Providing quick and accurate responses is crucial in telephone communication, but conventional methods make it difficult for callers to process large amounts of information instantly. Furthermore, callers need to shift their gaze to view display devices, which reduces the efficiency of the interaction. Additionally, generating responses tailored to specific industries or departments quickly is not easy. The objective of this invention is to solve these problems and improve the quality and efficiency of telephone communication.

[0956] 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.

[0957] In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the converted text and generating an appropriate response, means for displaying the generated response on a display device, and means for the display device to be integrated into a wearable device and to provide information to the call responder in real time. This allows the call responder to check the appropriate response in real time without moving their eyes and to respond immediately.

[0958] "Voice data" refers to the audio signals acquired during a phone call.

[0959] "Real-time" refers to processing that occurs at approximately the same rate as real-world time.

[0960] "Text" refers to the written information obtained by analyzing audio data.

[0961] "Conversion" refers to the process of converting audio data into text information.

[0962] "Analysis" refers to the process of understanding text data and extracting its meaning.

[0963] "Artificial intelligence tools" refer to algorithms and models that generate appropriate responses based on text data.

[0964] "Response" refers to an appropriate answer to a question or request, generated by artificial intelligence.

[0965] A "display device" refers to a device used to visually display the generated response.

[0966] A "wearable device" refers to a device that a user wears or uses on their body. Examples include smart glasses and head-mounted displays.

[0967] A "call handler" refers to a person who handles inquiries from customers or users.

[0968] "Providing information" refers to displaying the data and responses needed by the call operator on a display device in real time.

[0969] This invention is a system that improves the efficiency and accuracy of telephone communication. This system is characterized by converting acquired voice data into text in real time, analyzing the converted text to generate an appropriate response, and displaying the generated response on a display device of a wearable device.

[0970] Hardware and software to be used

[0971] Speech-to-Text API: Google Cloud Speech-to-Text

[0972] Generative artificial intelligence model: OpenAI GPT-4

[0973] Display device: Wearable device (e.g., smart glasses)

[0974] System configuration and operation

[0975] 1. Acquisition and conversion of audio data

[0976] The server converts the audio data acquired during a call into text in real time using the Google Cloud Speech-to-Text API. This ensures that the audio content of the call is instantly output as text.

[0977] 2. Text analysis and response generation

[0978] The server analyzes the converted text data and inputs it into OpenAI GPT-4. GPT-4, a generative AI model, generates appropriate responses based on pre-trained industry and sector knowledge.

[0979] Example of a prompt:

[0980] Please generate the best response to the following inquiry:

[0981] "My item hasn't arrived. What should I do?"

[0982] 3. Displaying the generated response

[0983] The generated response is sent from the server to the display device of a wearable device (smart glasses) and displayed for the caller to view in real time. This allows the caller to instantly grasp the information necessary to respond without having to shift their gaze significantly.

[0984] Specific example

[0985] scenario

[0986] 1. Call initiated (User)

[0987] A customer contacts a customer support representative wearing a wearable device and asks, "My product hasn't arrived. What should I do?"

[0988] 2. Speech recognition (server)

[0989] The server converts the customer's inquiry into text and generates the message, "My item hasn't arrived. What should I do?"

[0990] 3. Response generation (server)

[0991] The server inputs this text into a generative AI and generates the response, "We apologize for the inconvenience. If you could provide your order number, we will check it immediately."

[0992] 4. Response display (smart glasses)

[0993] The generated response is displayed on the smart glasses' screen, and the customer support representative reviews it.

[0994] 5. Final response (person in charge)

[0995] The staff member checks the response and replies to the customer, "We apologize for the inconvenience. Could you please tell us your order number?"

[0996] In this way, callers can check appropriate responses in real time through a wearable device without shifting their gaze, improving work efficiency and accuracy.

[0997] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0998] Step 1:

[0999] Acquisition of audio data

[1000] The user's call audio is captured through the microphone of the wearable device. The wearable device then sends this audio data to the server.

[1001] Input: User's call audio data

[1002] Output: Audio data sent to the server

[1003] Step 2:

[1004] Converting audio data to text

[1005] The server converts the audio data in real time using the Google Cloud Speech-to-Text API. In this process, the audio data is parsed via the API and the corresponding text data is output.

[1006] Input: Audio data

[1007] Output: Text data

[1008] Step 3:

[1009] Text data analysis

[1010] The server retrieves the converted text data and performs analysis to generate an appropriate response. This analysis understands the content of the text data and forms a response generation prompt based on that content.

[1011] Input: Text data

[1012] Output: Response generation prompt

[1013] Specific prompt message:

[1014] Please generate the best response to the following inquiry:

[1015] "My item hasn't arrived. What should I do?"

[1016] Step 4:

[1017] Response generation

[1018] The generative artificial intelligence model (OpenAI GPT-4) generates appropriate responses based on the generated prompts. In the response generation process, the generative AI model utilizes its pre-learned knowledge to produce highly accurate responses.

[1019] Input: Response generation prompt

[1020] Output: Generated response

[1021] Specific response:

[1022] "We apologize for the inconvenience. If you could provide your order number, we will check it immediately."

[1023] Step 5:

[1024] Sending the generated response

[1025] The generated response is sent from the server to the display device of the wearable device.

[1026] Input: Generated response

[1027] Output: Response data is sent to the wearable device.

[1028] Step 6:

[1029] Displaying the response

[1030] The wearable device displays the received response data on its screen. The response is displayed in real time so that the caller can see the content of the response.

[1031] Input: Response data

[1032] Output: The response displayed on the screen

[1033] Step 7:

[1034] Final response

[1035] The call operator will respond to the user appropriately based on the displayed response.

[1036] Input: Response displayed on the wearable device's screen.

[1037] Output: Final response to the user

[1038] Through this series of processing steps, users can respond to calls quickly and accurately.

[1039] 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.

[1040] This invention provides a system that improves efficiency and accuracy in telephone communication, and in particular, by combining it with a function that recognizes the user's emotions, it enables more appropriate responses. This system converts call audio into text in real time, inputs that text into a generative artificial intelligence (AI) to generate an appropriate response, and displays it to the caller. Furthermore, it incorporates an emotion engine that recognizes the user's emotions, and realizes technology that generates responses according to the emotional state.

[1041] System Configuration

[1042] This system mainly consists of the following four elements:

[1043] 1. Means for converting audio data into text (speech recognition engine)

[1044] 2. Response generation method using generative artificial intelligence (AI)

[1045] 3. Means for displaying the generated response (user interface)

[1046] 4. Means for recognizing the user's emotional state using an emotion engine

[1047] Acquisition and conversion of audio data (user and server)

[1048] The user communicates with the other party via telephone. During the call, audio data is transmitted in real time from the device to the server. The server uses a speech recognition engine to convert the audio data into text data. This process is performed using a speech recognition API.

[1049] Sentiment analysis (server)

[1050] The server uses an emotion engine to analyze the user's emotional state from the voice data. The emotion engine analyzes acoustic features such as tone, pitch, and speed of the voice to recognize the user's emotional state (e.g., stress, joy, anger, etc.). This recognition result is used in the subsequent response generation process.

[1051] Text analysis and response generation (server)

[1052] The server analyzes the text data converted by the speech recognition engine and inputs it into a generative artificial intelligence (AI). Based on pre-learned industry-specific and department-specific knowledge, the generative AI generates an appropriate response considering the text data and sentiment analysis results. This process enables flexible responses that respond to the user's emotions.

[1053] Display of the generated response (on the terminal)

[1054] The generated response is sent from the server to the user's terminal and displayed on the display device. The display device presents the information in a highly visible format that makes it easy for the user to confirm the response. This allows the call operator to immediately refer to the generated response.

[1055] Specific example

[1056] Here are some specific examples of its use:

[1057] Questions to ask when a user is feeling stressed.

[1058] 1. Call initiated (User)

[1059] The user receives a phone call, and the caller speaks in a strong tone, saying, "I am very dissatisfied with this service."

[1060] 2. Speech recognition and emotion recognition (server)

[1061] The server converts the speech to text and generates the text, "I am very dissatisfied with this service." At the same time, the emotion engine recognizes from the speech that the person is experiencing stress.

[1062] 3. Response generation (server)

[1063] The server passes this text and emotional state to the generative AI, which then generates a response such as, "I'm sorry. What part are you unhappy about?"

[1064] 4. Display of response (device)

[1065] The generated response is displayed on the user's monitor.

[1066] 5. Final interaction (user)

[1067] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "We apologize for the inconvenience. Specifically, what aspects are you dissatisfied with?"

[1068] Thus, by using this system, call operators can respond quickly and accurately while taking into account the user's emotional state. This is expected to improve the quality of service and increase user satisfaction.

[1069] The following describes the processing flow.

[1070] Step 1: Acquiring audio data (user)

[1071] The user initiates communication with the other party via telephone. Call audio data is captured in real time via the microphone.

[1072] Step 2: Transferring audio data (to the device)

[1073] The device sends the acquired voice data to the server. The voice data is transferred to the server in real time as data packets.

[1074] Step 3: Speech Recognition (Server)

[1075] The server receives the audio data and passes it to the speech recognition engine, which converts it into text data in real time. The speech recognition engine analyzes the audio data and converts it into text information.

[1076] Step 4: Sentiment Analysis (Server)

[1077] The server inputs the characteristics of the converted text and audio data into the emotion engine, which then analyzes the user's emotional state. The emotion engine recognizes the user's emotional state based on characteristics such as the tone, pitch, and speed of the voice.

[1078] Step 5: Text analysis and response generation (server)

[1079] The server inputs text data obtained through speech recognition and emotional information analyzed by the emotion engine into a generative artificial intelligence (AI) to generate an appropriate response. The generative AI utilizes industry-specific and departmental knowledge to derive the most appropriate response.

[1080] Step 6: Generating the response (server)

[1081] The generated responses are stored on the server as text data and adjusted as needed to match the user's emotional state.

[1082] Step 7: Send and view your reply (on your device)

[1083] The server sends the generated response to the terminal, which then displays it in real time on a display device connected to the terminal. The display device provides an intuitive and highly visible interface.

[1084] Step 8: Confirming the response and responding (to the user)

[1085] The user (the person answering the call) reviews the displayed response and makes minor adjustments to the response content as needed. By communicating the final response to the other party, an appropriate response is achieved that is tailored to the user's emotional state.

[1086] For example, if a user is feeling stressed and says, "I'm very dissatisfied with this service," the server converts the speech to text, the emotion engine recognizes the user's stress level, and the generative AI generates a response such as, "We are very sorry for the inconvenience. Specifically, what aspects are you dissatisfied with?" and displays it to the user. The user then refines the response by asking, "Specifically, what aspects are you dissatisfied with?" In this way, the system achieves more sophisticated call handling by combining emotion recognition.

[1087] (Example 2)

[1088] 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."

[1089] Traditional call answering systems struggled to provide timely responses that took into account the emotional state of the user during a call. As a result, communication between the call answerer and the user was often cumbersome, potentially leading to increased frustration and stress. Furthermore, it was difficult to leverage industry-specific and departmental expertise to generate appropriate responses, resulting in inconsistent response quality.

[1090] 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.

[1091] In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the emotional state of the converted text and voice and generating an appropriate response, and means for displaying the generated response on a display device of the call responder. This enables the provision of appropriate responses that reflect the emotional state of the user during a call in real time, thereby improving the quality of call handling and increasing user satisfaction.

[1092] "Acquired audio data" refers to audio information collected through the user's speech during a call.

[1093] "Real-time" refers to processing or conversion happening almost instantly with minimal delay.

[1094] "Means of converting to text" refers to technical devices or software for converting audio data into textual information.

[1095] "Converted text" refers to the textual information extracted from audio data.

[1096] "Voice emotional state" refers to the emotional characteristics and state of the user analyzed from their voice, such as stress, joy, or anger.

[1097] "Analyzing" refers to the process of examining data in detail to understand its meaning and state.

[1098] "An artificial intelligence method for generating appropriate responses" refers to artificial intelligence technology that creates the optimal response based on the user's utterance and emotional state.

[1099] "Generated response" refers to a response text to a user's utterance that is created by artificial intelligence.

[1100] A "call answerer" refers to a person whose role is to answer phone calls and conversations from users.

[1101] A "display device" refers to a device used to visually show generated responses or other information, such as a monitor or tablet.

[1102] "Means of display" refers to a technical device or software that outputs the generated response to a display device so that the caller can confirm it.

[1103] "Industry-specific and departmental knowledge" refers to specialized information and insights related to a particular industry or department.

[1104] "Prior learning" refers to a state where artificial intelligence has already acquired and understood the data and information necessary to generate a response.

[1105] "Means of adjustment" refers to technical devices or software that allow the call responder to modify or optimize the generated response.

[1106] Modes for carrying out the invention

[1107] This invention is a system that improves the efficiency and accuracy of call handling, and in particular, by combining it with a function that recognizes the user's emotions, it enables more appropriate responses. This system consists of the following four main elements:

[1108] 1. A means of converting acquired audio data into text in real time.

[1109] 2. An artificial intelligence means for analyzing the emotional state of converted text and audio and generating appropriate responses.

[1110] 3. Means for displaying the generated response on the caller's display device.

[1111] 4. A means by which the caller can adjust the displayed response.

[1112] This system uses a speech recognition engine, an emotion engine, a generative AI model, and a display device.

[1113] Acquisition and conversion of audio data (user and server)

[1114] Users communicate with others via telephone. Audio data from the call is transmitted in real time to a server by the device (e.g., a smartphone or telephone). The server converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text or IBM Watson Speech to Text). This process is performed using speech recognition APIs.

[1115] Sentiment analysis (server)

[1116] The server uses an emotion engine (e.g., Affectiva or Microsoft Azure Emotion API) to analyze the user's emotional state from the voice data. The emotion engine analyzes acoustic features such as tone, pitch, and speed of the voice to recognize the user's emotional state (e.g., stress, joy, anger, etc.). This recognition result is then used in the subsequent response generation process.

[1117] Text data analysis and response generation (server)

[1118] The server combines the converted text data with the sentiment analysis results and inputs them as prompts to a generative AI model (e.g., OpenAI GPT-4). For example, it might generate a prompt in the format of, "The user is feeling stressed. Please provide an appropriate response to 'I am very dissatisfied with this service.'" The generative AI model then generates an appropriate response based on this prompt.

[1119] Display and adjust the generated response (on the terminal)

[1120] The generated response is sent from the server to the user's device (e.g., a computer or tablet) and displayed on the device. The call operator can then review the displayed response and make adjustments as needed. For example, if the generated response is "I'm sorry, what exactly are you unhappy about?", they can adjust it to "I'm sorry for the inconvenience, but could you please elaborate on what specifically are you unhappy about?"

[1121] Specific example

[1122] Questions to ask when a user is feeling stressed.

[1123] 1. Call initiated (User)

[1124] The user receives a phone call, and the caller speaks in a strong tone, saying, "I am very dissatisfied with this service."

[1125] 2. Speech recognition and emotion recognition (server)

[1126] The server converts the speech to text and generates the text, "I am very dissatisfied with this service." At the same time, the emotion engine recognizes from the speech that the person is experiencing stress.

[1127] 3. Response generation (server)

[1128] The server passes this text and emotional state to the AI ​​model, which then generates a response such as, "I'm sorry. What part are you unhappy about?"

[1129] 4. Display of response (device)

[1130] The generated response is displayed on the user's monitor.

[1131] 5. Final interaction (user)

[1132] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "We apologize for the inconvenience. Specifically, what aspects are you dissatisfied with?"

[1133] This system allows call operators to respond quickly and accurately while considering the user's emotional state. This is expected to improve the quality of service and increase user satisfaction.

[1134] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1135] Step 1: Acquisition and transmission of audio data (user and device)

[1136] The user speaks over the phone. This audio data is captured by the device (e.g., a smartphone or telephone). The captured audio data is sent to the server in real time. The input is the user's voice, and the output is the audio data sent to the server.

[1137] Step 2: Convert audio data to text (server)

[1138] The server passes the received audio data to a speech recognition engine (for example, Google Cloud Speech-to-Text). The speech recognition engine converts the audio data into text data. At this stage, audio data is taken as input and text data is output. For example, the user's utterance is converted into the text "I am very dissatisfied with this service."

[1139] Step 3: Emotional state analysis (server)

[1140] The server passes the converted text data and raw audio data to an emotion engine (e.g., Microsoft Azure Emotion API). The emotion engine analyzes the user's emotional state from the audio data. At this stage, the input is the generated text data and audio data, and the output is the analysis result which includes the emotional state (e.g., the user is stressed).

[1141] Step 4: Generate prompt message (server)

[1142] The server generates prompts for the AI ​​model based on the acquired text data and emotional state. For example, it might generate a prompt such as, "The user is feeling stressed. Please provide an appropriate response to 'I am very dissatisfied with this service.'" The input is text data and emotional state, and the output is the generated prompt.

[1143] Step 5: Generating the response (server)

[1144] The server inputs a prompt into a generative AI model (e.g., OpenAI GPT-4). The generative AI model generates a response based on the prompt. The input is the prompt, and the output is an appropriate response to the user. For example, it might generate a response such as, "I'm sorry, but what part are you dissatisfied with?"

[1145] Step 6: Send the generated response (from server to terminal)

[1146] The generated response is sent from the server to the user's terminal. The input is the generated response, and the output is the data sent to the terminal.

[1147] Step 7: Display the generated response (on the terminal)

[1148] The generated response is displayed on a display device attached to the user's terminal (e.g., a monitor or tablet). The input is the response sent from the server, and the output is the text information displayed on the display device. Specifically, it might say, "We're sorry. What part are you dissatisfied with?"

[1149] Step 8: Final interaction (with the user)

[1150] The user reviews the displayed response and makes adjustments as needed. For example, the user might give a final response such as, "I'm sorry for the inconvenience. Specifically, what aspects are you dissatisfied with?" The input is the displayed response text, and the output is the user's final utterance.

[1151] (Application Example 2)

[1152] 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."

[1153] Traditional call answering systems often fail to consider the user's emotional state, leading to inadequate responses or inappropriate reactions. This can result in decreased user satisfaction and inconsistent service quality. Furthermore, it is difficult for call answerers to always provide the optimal response immediately, especially when dealing with emotionally charged users.

[1154] 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. In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the converted text and generating an appropriate response, emotion engine means for analyzing the user's emotional state, and means for displaying the response generated considering the emotion analysis results on the call operator's display device. This makes it possible to generate and provide an appropriate response to the call operator according to the user's emotional state.

[1155] "Voice data" refers to data that represents voice signals acquired during phone calls or voice communications in digital format.

[1156] "Real-time" refers to data processing and communication occurring almost instantly, resulting in minimal latency and instantaneous responsiveness.

[1157] "Text" refers to the string data that remains after audio data has been converted, and is expressed as written text in a form that humans can read.

[1158] "Artificial intelligence tools" are means of generating appropriate responses from input information using learning algorithms and data analysis techniques.

[1159] An "emotional engine" is a means of recognizing a user's emotional state by analyzing acoustic characteristics such as tone, pitch, and speed of speech.

[1160] A "call operator" refers to a person whose role is to respond to the other party through a phone call, and who uses the provided system or data to handle the call.

[1161] A "display device" is a device used to visually display generated responses or other information, and includes monitors and screens.

[1162] An embodiment of the present invention is a system that recognizes a user's emotions and generates and displays an optimal response according to that state. This system is particularly effective in user support for content distribution services, and responds quickly and accurately to user problems and questions through calls and chats.

[1163] System components

[1164] The system uses the following hardware and software:

[1165] hardware

[1166] Smartphone: Used as a device operated by the user, it acquires voice data and displays responses.

[1167] Server: A computer system used for analyzing voice data and generating responses.

[1168] software

[1169] Speech recognition APIs (e.g., Google Cloud Speech-to-Text): Convert speech data into text.

[1170] Emotion recognition engine (e.g., IBM Watson Tone Analyzer): Analyzes the user's emotional state from text.

[1171] Generative AI (e.g., OpenAI GPT-4): Generates appropriate responses based on the user's emotional state.

[1172] User Interface (UI): Displays the response generated on the smartphone or device.

[1173] System operation

[1174] The server uses a speech recognition API to convert audio data sent from the smartphone into text in real time. The converted text is then analyzed by an emotion recognition engine to identify the user's emotional state (e.g., stress, joy, anger). This emotion analysis is then used by a generative AI to generate an appropriate response. The generated response is sent from the server to the smartphone and displayed on the device.

[1175] Specific example

[1176] For example, if a user says, "The loading times are so slow, it's frustrating," the following process will occur:

[1177] 1. Acquisition and conversion of audio data

[1178] The smartphone's microphone captures audio, which is then sent to the server. The server uses the Google Cloud Speech-to-Text API to convert the audio data into text.

[1179] 2. Emotion analysis

[1180] The converted text is input into IBM Watson Tone Analyzer to recognize the user's emotional state (in this case, "frustration").

[1181] 3. Response generation

[1182] Based on the emotional state, OpenAI GPT-4 generates an appropriate response. For example, it might generate a response like, "We apologize for the inconvenience regarding loading times. We will address this immediately."

[1183] 4. Display the response

[1184] The generated response is sent from the server to the smartphone and displayed on the user interface.

[1185] Example of a prompt

[1186] User statement: "The loading times are so slow, it's frustrating."

[1187] User emotion: "Frustration"

[1188] Generate a supportive and empathetic response.

[1189] By using this system, it becomes possible to respond appropriately while taking user emotions into consideration, and it is expected that this will improve user satisfaction with content distribution services.

[1190] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1191] Program processing steps

[1192] Step 1:

[1193] The user uses their smartphone's microphone to input voice data. The smartphone acquires this voice data and sends it to the server in real time.

[1194] Input: User's voice

[1195] Output: Audio data sent to the server

[1196] Step 2:

[1197] The server inputs the received audio data into a speech recognition API and converts it into text data. For example, Google Cloud Speech-to-Text is used as the speech recognition API.

[1198] Input: Audio data sent to the server

[1199] Output: Text data

[1200] Step 3:

[1201] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotional state. IBM Watson Tone Analyzer is used as the emotion recognition engine.

[1202] Input: Text data

[1203] Output: User's emotional state (e.g., "frustration")

[1204] Step 4:

[1205] The server inputs the analyzed emotional state and text data into a generative AI to generate an appropriate response. OpenAI GPT-4 is used as the generative AI.

[1206] Input: Text data and user's emotional state

[1207] Output: Appropriate response

[1208] Step 5:

[1209] The server sends the generated response to the smartphone. The smartphone's user interface displays the received response.

[1210] Input: Generated response

[1211] Output: Response displayed on the smartphone's display screen

[1212] Specific operation of the process

[1213] Specific actions for Step 1:

[1214] When a user says to their smartphone, "The loading times are so slow, it's frustrating," the smartphone's microphone captures this audio and sends it to the server.

[1215] Specific actions in Step 2:

[1216] The server sends audio data to the Google Cloud Speech-to-Text API and receives this audio data as text. For example, the speech recognition API outputs the text "The loading time is so slow, it's frustrating."

[1217] Specific actions in Step 3:

[1218] The server sends the text message "The loading time is so slow, it's frustrating" to IBM Watson Tone Analyzer, and the analysis results in an emotional state of "frustration."

[1219] Specific actions in Step 4:

[1220] The server inputs the text data "The loading time is slow and frustrating" and the emotion state "frustrated" into OpenAI GPT-4 and generates a response such as "We apologize. We understand that you are experiencing inconvenience with the loading time. We will address this immediately."

[1221] Specific actions in Step 5:

[1222] The server sends the generated response to the smartphone. The smartphone's user interface then displays this response visually to the user.

[1223] In this way, a system can be realized that takes into account the user's emotions in real time and generates and displays appropriate responses.

[1224] 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.

[1225] 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.

[1226] 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.

[1227] [Fourth Embodiment]

[1228] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[1229] 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.

[1230] 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).

[1231] 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.

[1232] 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.

[1233] 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).

[1234] 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.

[1235] 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.

[1236] 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.

[1237] 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.

[1238] 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.

[1239] 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.

[1240] 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".

[1241] This invention provides a system that improves the efficiency and accuracy of telephone communication. This system realizes a technology that converts telephone audio into text in real time, inputs that text data into a generative artificial intelligence (AI) to generate an appropriate response, and displays it to the person handling the call.

[1242] System Configuration

[1243] This system mainly consists of the following three elements:

[1244] 1. Means for converting audio data into text (speech recognition engine)

[1245] 2. Response generation method using generative artificial intelligence (AI)

[1246] 3. Means for displaying response results (user interface)

[1247] Acquisition and conversion of audio data (server)

[1248] The server converts the audio data acquired during a call into text data in real time. The speech recognition engine can analyze the audio data and convert it into text information. In this process, the server uses a speech recognition API to output the audio data as text data.

[1249] Text analysis and response generation (server)

[1250] After text recognition processing, the server analyzes the recognized text data and inputs it into a generative AI. The generative AI generates appropriate responses to questions based on pre-learned industry and departmental knowledge. This AI utilizes state-of-the-art machine learning algorithms to provide highly accurate responses.

[1251] Display of the generated response (on the terminal)

[1252] The generated response is sent from the server to the user's terminal and displayed on the terminal's display. Because the response is displayed in real time for the call operator to review, the user can refer to it immediately.

[1253] Specific example

[1254] Here are some specific examples of its use:

[1255] Questions about registration

[1256] 1. Call initiated (User)

[1257] The user receives a phone call, and the caller asks, "How do I register for a new account?"

[1258] 2. Speech recognition (server)

[1259] The server converts the audio to text and generates the text, "Please tell me how to register for a new account."

[1260] 3. Response generation (server)

[1261] The server passes this text to the generative AI, which then generates a response saying, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[1262] 4. Display of response (device)

[1263] The generated response is displayed on the user's monitor.

[1264] 5. Final interaction (user)

[1265] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "First, please visit our website and click the 'Register' button in the upper right corner. Then, please fill in the required information."

[1266] In this way, this system enables call operators to provide quick and accurate responses. By referring to responses generated by generative AI, call operators can immediately utilize their specialized knowledge, thereby improving the quality of their responses.

[1267] The following describes the processing flow.

[1268] Step 1: Acquiring audio data (user)

[1269] The user initiates communication with the other party via telephone. Voice data is acquired in real time through the microphone.

[1270] Step 2: Transferring audio data (to the device)

[1271] The terminal sends the acquired voice data to the server. The voice data is transferred in real time as data packets.

[1272] Step 3: Speech Recognition (Server)

[1273] The server receives the audio data and passes it to the speech recognition engine, which converts it into text data in real time. The speech recognition engine analyzes the audio data and generates corresponding text information.

[1274] Step 4: Text analysis and response generation (server)

[1275] The server analyzes the converted text data to understand its context and content. It then inputs this text into a generative artificial intelligence (AI) to generate an appropriate response. The generative AI utilizes pre-learned industry- and department-specific knowledge to produce highly accurate responses.

[1276] Step 5: Generating the response (server)

[1277] The generated response is reviewed and sent to the caller. The response is sent from the server to the terminal as text data.

[1278] Step 6: Displaying the response (on the device)

[1279] The terminal displays the received response on the caller's monitor. The display device shows the response in a highly visible and easy-to-understand format.

[1280] Step 7: Confirmation of response and response (user)

[1281] The user (call operator) reviews the displayed response. If necessary, the user adjusts the response and provides an appropriate final response to the other party. This allows the call operator to provide flexible and accurate responses while referring to the responses generated by the AI.

[1282] (Example 1)

[1283] 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".

[1284] Traditional call answering systems struggled to convert call audio into text in real time and generate appropriate responses. Furthermore, they lacked systems that enabled callers to instantly utilize their specialized knowledge to provide quick and accurate responses. As a result, the efficiency and accuracy of responses decreased, making it difficult to improve customer satisfaction.

[1285] 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.

[1286] In this invention, the server includes means for converting acquired audio data into text in real time, means for analyzing the converted text and inputting it as a prompt sentence into a generative AI model that has been pre-trained with industry-specific and department-specific knowledge, and means for displaying the response generated by the generative AI model on the call operator's display device. This makes it possible to instantly transcribe call audio into text and generate and display an appropriate response.

[1287] "Voice data" refers to digital signals obtained from voice, and is data used to record the content of a conversation in real time.

[1288] "Converting to text" refers to the process of analyzing acquired audio data and converting the linguistic information into string-based data.

[1289] A "generative AI model" is an artificial intelligence system that uses deep learning technology to dynamically generate appropriate responses to newly input text based on previously learned knowledge.

[1290] A "prompt sentence" is the initial text data input to a generative AI model, and it is the sentence that forms the basis for generating a response.

[1291] A "call operator" refers to a person who communicates with customers using a call system.

[1292] A "display device" is hardware used to visually display generated responses or other information, and includes monitors, displays, and screens.

[1293] "Real-time" refers to a state where processing and responses occur almost instantly, with virtually no delay or time lag.

[1294] "User interface means" refers to interactive screens or components that allow the call responder to review and adjust the generated response.

[1295] "Industry-specific and department-specific knowledge" refers to specialized information and knowledge related to a particular industry or department that has been learned in advance, forming the foundation for accurate and appropriate responses.

[1296] This invention provides a system that improves efficiency and accuracy in telephone communication. This system realizes a technology that converts telephone audio into text in real time, inputs that text data into a generating AI model to generate an appropriate response, and displays it to the person handling the call.

[1297] Acquisition and conversion of audio data

[1298] The server converts audio data acquired during a call into text data in real time. This process uses a speech recognition engine such as the Google Cloud Speech-to-Text API to analyze the audio data and convert the linguistic information into text data. For example, if a user says, "Please tell me how to register for a new account," the server converts this audio data into the text "Please tell me how to register for a new account."

[1299] Text analysis and response generation

[1300] The server analyzes the acquired text data, identifies key keywords and context, and then inputs it into a generative AI model. Examples of generative AI models used here include OpenAI GPT-3 and ChatGPT. This model leverages pre-trained industry- and department-specific knowledge to generate appropriate responses based on the text data provided as prompts. For example, in response to the prompt "How do I register for a new account?", it would generate a response such as "To register for a new account, first visit our website and click the 'Register' button in the upper right corner."

[1301] Display of the generated response

[1302] The generated response is sent from the server to the user's terminal. The terminal has a web browser or dedicated application installed, which visually displays the received response on a display device. The display format and layout follow a pre-configured template, allowing the call operator to quickly review the response. The displayed response can also be adjusted via the user interface as needed.

[1303] Specific example

[1304] As a concrete example of its use, consider the following scenario:

[1305] 1. Call initiated (User)

[1306] The user receives a phone call and is asked, "How do I register for a new account?"

[1307] 2. Speech recognition (server)

[1308] The server converts the audio to text and generates the text, "Please tell me how to register for a new account."

[1309] 3. Response generation (server)

[1310] The server passes this text to the AI ​​model, which then generates the response, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[1311] 4. Display of response (device)

[1312] The generated response is displayed on the user's monitor.

[1313] 5. Final interaction (user)

[1314] The caller will review the displayed text and, if necessary, provide a final response such as, "First, please visit our website and click the 'Register' button in the upper right corner. Then, please fill in the required information."

[1315] In this way, this system enables call operators to provide quick and accurate responses. By referring to responses generated by the AI ​​model, call operators can immediately utilize their specialized knowledge, thereby improving the quality of their responses.

[1316] The flow of the specific processing in Example 1 will be explained using Figure 11.

[1317] System program processing flow

[1318] Step 1:

[1319] Acquisition of audio data

[1320] The server acquires voice data from the call system in real time. When a user says, "Please tell me how to register a new account," this voice data is sent to the server from the VoIP system or telephone system.

[1321] Input: Voice data from the call system

[1322] Output: Audio data file

[1323] Specific operation: The server receives the audio data and temporarily stores it.

[1324] Step 2:

[1325] Converting audio data to text

[1326] The server converts the acquired audio data into text data in real time. This process uses a speech recognition engine (for example, the Google Cloud Speech-to-Text API). The audio data is analyzed, and the text "How do I register for a new account?" is generated.

[1327] Input: Audio data

[1328] Output: Text data

[1329] Specific operation: The server calls a speech recognition API and outputs the audio data as text.

[1330] Step 3:

[1331] Text data analysis

[1332] The server analyzes the converted text data to identify key keywords and context. Based on this analysis, it prepares the data to be input as prompts to the generative AI model.

[1333] Input: Text data

[1334] Output: Analyzed text data

[1335] Specific operation: The server uses a parsing algorithm to analyze the structure and meaning of the text and format it into an appropriate prompt message.

[1336] Step 4:

[1337] Response generation using a generative AI model

[1338] The server inputs the analyzed text data into a generative AI model. This generative AI model (for example, OpenAI GPT-3) uses its pre-trained knowledge to generate an appropriate response such as, "To register a new account, first visit our website and click the 'Register' button in the upper right corner."

[1339] Input: Analyzed text data

[1340] Output: Generated response

[1341] Specific operation: The server sends a prompt to the generated AI model and receives the response in text format.

[1342] Step 5:

[1343] Preparing to display the generated response

[1344] The server receives the generated response and prepares to send it directly to the display device without going through the user interface. Display formatting and layout are applied, and the response is formatted in a way that is suitable for the user to read.

[1345] Input: Generated response

[1346] Output: Formatted text data for display

[1347] Specific operation: The server appropriately formats the response text and formats it according to the user interface template.

[1348] Step 6:

[1349] Display of the generated response

[1350] The server sends the formatted response data to the user's terminal. The terminal receives this data and displays it visually on its display device. The user (the person answering the call) reviews the displayed response and makes adjustments as needed.

[1351] Input: Formatted response data

[1352] Output: Response displayed on the display device

[1353] Specific operation: The server sends the formatted response data to the terminal, which then displays it on its display device. The user confirms this and verbally communicates it to the person they are talking to.

[1354] Through the above processing steps, it becomes possible to transcribe call audio into text in real time, generate appropriate responses using a generation AI model, and provide callers with quick and accurate information.

[1355] (Application Example 1)

[1356] 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".

[1357] Providing quick and accurate responses is crucial in telephone communication, but conventional methods make it difficult for callers to process large amounts of information instantly. Furthermore, callers need to shift their gaze to view display devices, which reduces the efficiency of the interaction. Additionally, generating responses tailored to specific industries or departments quickly is not easy. The objective of this invention is to solve these problems and improve the quality and efficiency of telephone communication.

[1358] 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.

[1359] In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the converted text and generating an appropriate response, means for displaying the generated response on a display device, and means for the display device to be integrated into a wearable device and to provide information to the call responder in real time. This allows the call responder to check the appropriate response in real time without moving their eyes and to respond immediately.

[1360] "Voice data" refers to the audio signals acquired during a phone call.

[1361] "Real-time" refers to processing that occurs at approximately the same rate as real-world time.

[1362] "Text" refers to the written information obtained by analyzing audio data.

[1363] "Conversion" refers to the process of converting audio data into text information.

[1364] "Analysis" refers to the process of understanding text data and extracting its meaning.

[1365] "Artificial intelligence tools" refer to algorithms and models that generate appropriate responses based on text data.

[1366] "Response" refers to an appropriate answer to a question or request, generated by artificial intelligence.

[1367] A "display device" refers to a device used to visually display the generated response.

[1368] A "wearable device" refers to a device that a user wears or uses on their body. Examples include smart glasses and head-mounted displays.

[1369] A "call handler" refers to a person who handles inquiries from customers or users.

[1370] "Providing information" refers to displaying the data and responses needed by the call operator on a display device in real time.

[1371] This invention is a system that improves the efficiency and accuracy of telephone communication. This system is characterized by converting acquired voice data into text in real time, analyzing the converted text to generate an appropriate response, and displaying the generated response on a display device of a wearable device.

[1372] Hardware and software to be used

[1373] Speech-to-Text API: Google Cloud Speech-to-Text

[1374] Generative artificial intelligence model: OpenAI GPT-4

[1375] Display device: Wearable device (e.g., smart glasses)

[1376] System configuration and operation

[1377] 1. Acquisition and conversion of audio data

[1378] The server converts the audio data acquired during a call into text in real time using the Google Cloud Speech-to-Text API. This ensures that the audio content of the call is instantly output as text.

[1379] 2. Text analysis and response generation

[1380] The server analyzes the converted text data and inputs it into OpenAI GPT-4. GPT-4, a generative AI model, generates appropriate responses based on pre-trained industry and sector knowledge.

[1381] Example of a prompt:

[1382] Please generate the best response to the following inquiry:

[1383] "My item hasn't arrived. What should I do?"

[1384] 3. Displaying the generated response

[1385] The generated response is sent from the server to the display device of a wearable device (smart glasses) and displayed for the caller to view in real time. This allows the caller to instantly grasp the information necessary to respond without having to shift their gaze significantly.

[1386] Specific example

[1387] scenario

[1388] 1. Call initiated (User)

[1389] A customer contacts a customer support representative wearing a wearable device and asks, "My product hasn't arrived. What should I do?"

[1390] 2. Speech recognition (server)

[1391] The server converts the customer's inquiry into text and generates the message, "My item hasn't arrived. What should I do?"

[1392] 3. Response generation (server)

[1393] The server inputs this text into a generative AI and generates the response, "We apologize for the inconvenience. If you could provide your order number, we will check it immediately."

[1394] 4. Response display (smart glasses)

[1395] The generated response is displayed on the smart glasses' screen, and the customer support representative reviews it.

[1396] 5. Final response (person in charge)

[1397] The staff member checks the response and replies to the customer, "We apologize for the inconvenience. Could you please tell us your order number?"

[1398] In this way, callers can check appropriate responses in real time through a wearable device without shifting their gaze, improving work efficiency and accuracy.

[1399] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[1400] Step 1:

[1401] Acquisition of audio data

[1402] The user's call audio is captured through the microphone of the wearable device. The wearable device then sends this audio data to the server.

[1403] Input: User's call audio data

[1404] Output: Audio data sent to the server

[1405] Step 2:

[1406] Converting audio data to text

[1407] The server converts the audio data in real time using the Google Cloud Speech-to-Text API. In this process, the audio data is parsed via the API and the corresponding text data is output.

[1408] Input: Audio data

[1409] Output: Text data

[1410] Step 3:

[1411] Text data analysis

[1412] The server retrieves the converted text data and performs analysis to generate an appropriate response. This analysis understands the content of the text data and forms a response generation prompt based on that content.

[1413] Input: Text data

[1414] Output: Response generation prompt

[1415] Specific prompt message:

[1416] Please generate the best response to the following inquiry:

[1417] "My item hasn't arrived. What should I do?"

[1418] Step 4:

[1419] Response generation

[1420] The generative artificial intelligence model (OpenAI GPT-4) generates appropriate responses based on the generated prompts. In the response generation process, the generative AI model utilizes its pre-learned knowledge to produce highly accurate responses.

[1421] Input: Response generation prompt

[1422] Output: Generated response

[1423] Specific response:

[1424] "We apologize for the inconvenience. If you could provide your order number, we will check it immediately."

[1425] Step 5:

[1426] Sending the generated response

[1427] The generated response is sent from the server to the display device of the wearable device.

[1428] Input: Generated response

[1429] Output: Response data is sent to the wearable device.

[1430] Step 6:

[1431] Displaying the response

[1432] The wearable device displays the received response data on its screen. The response is displayed in real time so that the caller can see the content of the response.

[1433] Input: Response data

[1434] Output: The response displayed on the screen

[1435] Step 7:

[1436] Final response

[1437] The call operator will respond to the user appropriately based on the displayed response.

[1438] Input: Response displayed on the wearable device's screen.

[1439] Output: Final response to the user

[1440] Through this series of processing steps, users can respond to calls quickly and accurately.

[1441] 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.

[1442] This invention provides a system that improves efficiency and accuracy in telephone communication, and in particular, by combining it with a function that recognizes the user's emotions, it enables more appropriate responses. This system converts call audio into text in real time, inputs that text into a generative artificial intelligence (AI) to generate an appropriate response, and displays it to the caller. Furthermore, it incorporates an emotion engine that recognizes the user's emotions, and realizes technology that generates responses according to the emotional state.

[1443] System Configuration

[1444] This system mainly consists of the following four elements:

[1445] 1. Means for converting audio data into text (speech recognition engine)

[1446] 2. Response generation method using generative artificial intelligence (AI)

[1447] 3. Means for displaying the generated response (user interface)

[1448] 4. Means for recognizing the user's emotional state using an emotion engine

[1449] Acquisition and conversion of audio data (user and server)

[1450] The user communicates with the other party via telephone. During the call, audio data is transmitted in real time from the device to the server. The server uses a speech recognition engine to convert the audio data into text data. This process is performed using a speech recognition API.

[1451] Sentiment analysis (server)

[1452] The server uses an emotion engine to analyze the user's emotional state from the voice data. The emotion engine analyzes acoustic features such as tone, pitch, and speed of the voice to recognize the user's emotional state (e.g., stress, joy, anger, etc.). This recognition result is used in the subsequent response generation process.

[1453] Text analysis and response generation (server)

[1454] The server analyzes the text data converted by the speech recognition engine and inputs it into a generative artificial intelligence (AI). Based on pre-learned industry-specific and department-specific knowledge, the generative AI generates an appropriate response considering the text data and sentiment analysis results. This process enables flexible responses that respond to the user's emotions.

[1455] Display of the generated response (on the terminal)

[1456] The generated response is sent from the server to the user's terminal and displayed on the display device. The display device presents the information in a highly visible format that makes it easy for the user to confirm the response. This allows the call operator to immediately refer to the generated response.

[1457] Specific example

[1458] Here are some specific examples of its use:

[1459] Questions to ask when a user is feeling stressed.

[1460] 1. Call initiated (User)

[1461] The user receives a phone call, and the caller speaks in a strong tone, saying, "I am very dissatisfied with this service."

[1462] 2. Speech recognition and emotion recognition (server)

[1463] The server converts the speech to text and generates the text, "I am very dissatisfied with this service." At the same time, the emotion engine recognizes from the speech that the person is experiencing stress.

[1464] 3. Response generation (server)

[1465] The server passes this text and emotional state to the generative AI, which then generates a response such as, "I'm sorry. What part are you unhappy about?"

[1466] 4. Display of response (device)

[1467] The generated response is displayed on the user's monitor.

[1468] 5. Final interaction (user)

[1469] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "We apologize for the inconvenience. Specifically, what aspects are you dissatisfied with?"

[1470] Thus, by using this system, call operators can respond quickly and accurately while taking into account the user's emotional state. This is expected to improve the quality of service and increase user satisfaction.

[1471] The following describes the processing flow.

[1472] Step 1: Acquiring audio data (user)

[1473] The user initiates communication with the other party via telephone. Call audio data is captured in real time via the microphone.

[1474] Step 2: Transferring audio data (to the device)

[1475] The device sends the acquired voice data to the server. The voice data is transferred to the server in real time as data packets.

[1476] Step 3: Speech Recognition (Server)

[1477] The server receives the audio data and passes it to the speech recognition engine, which converts it into text data in real time. The speech recognition engine analyzes the audio data and converts it into text information.

[1478] Step 4: Sentiment Analysis (Server)

[1479] The server inputs the characteristics of the converted text and audio data into the emotion engine, which then analyzes the user's emotional state. The emotion engine recognizes the user's emotional state based on characteristics such as the tone, pitch, and speed of the voice.

[1480] Step 5: Text analysis and response generation (server)

[1481] The server inputs text data obtained through speech recognition and emotional information analyzed by the emotion engine into a generative artificial intelligence (AI) to generate an appropriate response. The generative AI utilizes industry-specific and departmental knowledge to derive the most appropriate response.

[1482] Step 6: Generating the response (server)

[1483] The generated responses are stored on the server as text data and adjusted as needed to match the user's emotional state.

[1484] Step 7: Send and view your reply (on your device)

[1485] The server sends the generated response to the terminal, which then displays it in real time on a display device connected to the terminal. The display device provides an intuitive and highly visible interface.

[1486] Step 8: Confirming the response and responding (to the user)

[1487] The user (the person answering the call) reviews the displayed response and makes minor adjustments to the response content as needed. By communicating the final response to the other party, an appropriate response is achieved that is tailored to the user's emotional state.

[1488] For example, if a user is feeling stressed and says, "I'm very dissatisfied with this service," the server converts the speech to text, the emotion engine recognizes the user's stress level, and the generative AI generates a response such as, "We are very sorry for the inconvenience. Specifically, what aspects are you dissatisfied with?" and displays it to the user. The user then refines the response by asking, "Specifically, what aspects are you dissatisfied with?" In this way, the system achieves more sophisticated call handling by combining emotion recognition.

[1489] (Example 2)

[1490] 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".

[1491] Traditional call answering systems struggled to provide timely responses that took into account the emotional state of the user during a call. As a result, communication between the call answerer and the user was often cumbersome, potentially leading to increased frustration and stress. Furthermore, it was difficult to leverage industry-specific and departmental expertise to generate appropriate responses, resulting in inconsistent response quality.

[1492] 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.

[1493] In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the emotional state of the converted text and voice and generating an appropriate response, and means for displaying the generated response on a display device of the call responder. This enables the provision of appropriate responses that reflect the emotional state of the user during a call in real time, thereby improving the quality of call handling and increasing user satisfaction.

[1494] "Acquired audio data" refers to audio information collected through the user's speech during a call.

[1495] "Real-time" refers to processing or conversion happening almost instantly with minimal delay.

[1496] "Means of converting to text" refers to technical devices or software for converting audio data into textual information.

[1497] "Converted text" refers to the textual information extracted from audio data.

[1498] "Voice emotional state" refers to the emotional characteristics and state of the user analyzed from their voice, such as stress, joy, or anger.

[1499] "Analyzing" refers to the process of examining data in detail to understand its meaning and state.

[1500] "An artificial intelligence method for generating appropriate responses" refers to artificial intelligence technology that creates the optimal response based on the user's utterance and emotional state.

[1501] "Generated response" refers to a response text to a user's utterance that is created by artificial intelligence.

[1502] A "call answerer" refers to a person whose role is to answer phone calls and conversations from users.

[1503] A "display device" refers to a device used to visually show generated responses or other information, such as a monitor or tablet.

[1504] "Means of display" refers to a technical device or software that outputs the generated response to a display device so that the caller can confirm it.

[1505] "Industry-specific and departmental knowledge" refers to specialized information and insights related to a particular industry or department.

[1506] "Prior learning" refers to a state where artificial intelligence has already acquired and understood the data and information necessary to generate a response.

[1507] "Means of adjustment" refers to technical devices or software that allow the call responder to modify or optimize the generated response.

[1508] Modes for carrying out the invention

[1509] This invention is a system that improves the efficiency and accuracy of call handling, and in particular, by combining it with a function that recognizes the user's emotions, it enables more appropriate responses. This system consists of the following four main elements:

[1510] 1. A means of converting acquired audio data into text in real time.

[1511] 2. An artificial intelligence means for analyzing the emotional state of converted text and audio and generating appropriate responses.

[1512] 3. Means for displaying the generated response on the caller's display device.

[1513] 4. A means by which the caller can adjust the displayed response.

[1514] This system uses a speech recognition engine, an emotion engine, a generative AI model, and a display device.

[1515] Acquisition and conversion of audio data (user and server)

[1516] Users communicate with others via telephone. Audio data from the call is transmitted in real time to a server by the device (e.g., a smartphone or telephone). The server converts the audio data into text data using a speech recognition engine (e.g., Google Cloud Speech-to-Text or IBM Watson Speech to Text). This process is performed using speech recognition APIs.

[1517] Sentiment analysis (server)

[1518] The server uses an emotion engine (e.g., Affectiva or Microsoft Azure Emotion API) to analyze the user's emotional state from the voice data. The emotion engine analyzes acoustic features such as tone, pitch, and speed of the voice to recognize the user's emotional state (e.g., stress, joy, anger, etc.). This recognition result is then used in the subsequent response generation process.

[1519] Text data analysis and response generation (server)

[1520] The server combines the converted text data with the sentiment analysis results and inputs them as prompts to a generative AI model (e.g., OpenAI GPT-4). For example, it might generate a prompt in the format of, "The user is feeling stressed. Please provide an appropriate response to 'I am very dissatisfied with this service.'" The generative AI model then generates an appropriate response based on this prompt.

[1521] Display and adjust the generated response (on the terminal)

[1522] The generated response is sent from the server to the user's device (e.g., a computer or tablet) and displayed on the device. The call operator can then review the displayed response and make adjustments as needed. For example, if the generated response is "I'm sorry, what exactly are you unhappy about?", they can adjust it to "I'm sorry for the inconvenience, but could you please elaborate on what specifically are you unhappy about?"

[1523] Specific example

[1524] Questions to ask when a user is feeling stressed.

[1525] 1. Call initiated (User)

[1526] The user receives a phone call, and the caller speaks in a strong tone, saying, "I am very dissatisfied with this service."

[1527] 2. Speech recognition and emotion recognition (server)

[1528] The server converts the speech to text and generates the text, "I am very dissatisfied with this service." At the same time, the emotion engine recognizes from the speech that the person is experiencing stress.

[1529] 3. Response generation (server)

[1530] The server passes this text and emotional state to the AI ​​model, which then generates a response such as, "I'm sorry. What part are you unhappy about?"

[1531] 4. Display of response (device)

[1532] The generated response is displayed on the user's monitor.

[1533] 5. Final interaction (user)

[1534] The user reviews the displayed text, makes adjustments as needed, and then provides a final response such as, "We apologize for the inconvenience. Specifically, what aspects are you dissatisfied with?"

[1535] This system allows call operators to respond quickly and accurately while considering the user's emotional state. This is expected to improve the quality of service and increase user satisfaction.

[1536] The flow of the specific processing in Example 2 will be explained using Figure 13.

[1537] Step 1: Acquisition and transmission of audio data (user and device)

[1538] The user speaks over the phone. This audio data is captured by the device (e.g., a smartphone or telephone). The captured audio data is sent to the server in real time. The input is the user's voice, and the output is the audio data sent to the server.

[1539] Step 2: Convert audio data to text (server)

[1540] The server passes the received audio data to a speech recognition engine (for example, Google Cloud Speech-to-Text). The speech recognition engine converts the audio data into text data. At this stage, audio data is taken as input and text data is output. For example, the user's utterance is converted into the text "I am very dissatisfied with this service."

[1541] Step 3: Emotional state analysis (server)

[1542] The server passes the converted text data and raw audio data to an emotion engine (e.g., Microsoft Azure Emotion API). The emotion engine analyzes the user's emotional state from the audio data. At this stage, the input is the generated text data and audio data, and the output is the analysis result which includes the emotional state (e.g., the user is stressed).

[1543] Step 4: Generate prompt message (server)

[1544] The server generates prompts for the AI ​​model based on the acquired text data and emotional state. For example, it might generate a prompt such as, "The user is feeling stressed. Please provide an appropriate response to 'I am very dissatisfied with this service.'" The input is text data and emotional state, and the output is the generated prompt.

[1545] Step 5: Generating the response (server)

[1546] The server inputs a prompt into a generative AI model (e.g., OpenAI GPT-4). The generative AI model generates a response based on the prompt. The input is the prompt, and the output is an appropriate response to the user. For example, it might generate a response such as, "I'm sorry, but what part are you dissatisfied with?"

[1547] Step 6: Send the generated response (from server to terminal)

[1548] The generated response is sent from the server to the user's terminal. The input is the generated response, and the output is the data sent to the terminal.

[1549] Step 7: Display the generated response (on the terminal)

[1550] The generated response is displayed on a display device attached to the user's terminal (e.g., a monitor or tablet). The input is the response sent from the server, and the output is the text information displayed on the display device. Specifically, it might say, "We're sorry. What part are you dissatisfied with?"

[1551] Step 8: Final interaction (with the user)

[1552] The user reviews the displayed response and makes adjustments as needed. For example, the user might give a final response such as, "I'm sorry for the inconvenience. Specifically, what aspects are you dissatisfied with?" The input is the displayed response text, and the output is the user's final utterance.

[1553] (Application Example 2)

[1554] 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".

[1555] Traditional call answering systems often fail to consider the user's emotional state, leading to inadequate responses or inappropriate reactions. This can result in decreased user satisfaction and inconsistent service quality. Furthermore, it is difficult for call answerers to always provide the optimal response immediately, especially when dealing with emotionally charged users.

[1556] 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. In this invention, the server includes means for converting acquired voice data into text in real time, artificial intelligence means for analyzing the converted text and generating an appropriate response, emotion engine means for analyzing the user's emotional state, and means for displaying the response generated considering the emotion analysis results on the call operator's display device. This makes it possible to generate and provide an appropriate response to the call operator according to the user's emotional state.

[1557] "Voice data" refers to data that represents voice signals acquired during phone calls or voice communications in digital format.

[1558] "Real-time" refers to data processing and communication occurring almost instantly, resulting in minimal latency and instantaneous responsiveness.

[1559] "Text" refers to the string data that remains after audio data has been converted, and is expressed as written text in a form that humans can read.

[1560] "Artificial intelligence tools" are means of generating appropriate responses from input information using learning algorithms and data analysis techniques.

[1561] An "emotional engine" is a means of recognizing a user's emotional state by analyzing acoustic characteristics such as tone, pitch, and speed of speech.

[1562] A "call operator" refers to a person whose role is to respond to the other party through a phone call, and who uses the provided system or data to handle the call.

[1563] A "display device" is a device used to visually display generated responses or other information, and includes monitors and screens.

[1564] An embodiment of the present invention is a system that recognizes a user's emotions and generates and displays an optimal response according to that state. This system is particularly effective in user support for content distribution services, and responds quickly and accurately to user problems and questions through calls and chats.

[1565] System components

[1566] The system uses the following hardware and software:

[1567] hardware

[1568] Smartphone: Used as a device operated by the user, it acquires voice data and displays responses.

[1569] Server: A computer system used for analyzing voice data and generating responses.

[1570] software

[1571] Speech recognition APIs (e.g., Google Cloud Speech-to-Text): Convert speech data into text.

[1572] Emotion recognition engine (e.g., IBM Watson Tone Analyzer): Analyzes the user's emotional state from text.

[1573] Generative AI (e.g., OpenAI GPT-4): Generates appropriate responses based on the user's emotional state.

[1574] User Interface (UI): Displays the response generated on the smartphone or device.

[1575] System operation

[1576] The server uses a speech recognition API to convert audio data sent from the smartphone into text in real time. The converted text is then analyzed by an emotion recognition engine to identify the user's emotional state (e.g., stress, joy, anger). This emotion analysis is then used by a generative AI to generate an appropriate response. The generated response is sent from the server to the smartphone and displayed on the device.

[1577] Specific example

[1578] For example, if a user says, "The loading times are so slow, it's frustrating," the following process will occur:

[1579] 1. Acquisition and conversion of audio data

[1580] The smartphone's microphone captures audio, which is then sent to the server. The server uses the Google Cloud Speech-to-Text API to convert the audio data into text.

[1581] 2. Emotion analysis

[1582] The converted text is input into IBM Watson Tone Analyzer to recognize the user's emotional state (in this case, "frustration").

[1583] 3. Response generation

[1584] Based on the emotional state, OpenAI GPT-4 generates an appropriate response. For example, it might generate a response like, "We apologize for the inconvenience regarding loading times. We will address this immediately."

[1585] 4. Display the response

[1586] The generated response is sent from the server to the smartphone and displayed on the user interface.

[1587] Example of a prompt

[1588] User statement: "The loading times are so slow, it's frustrating."

[1589] User emotion: "Frustration"

[1590] Generate a supportive and empathetic response.

[1591] By using this system, it becomes possible to respond appropriately while taking user emotions into consideration, and it is expected that this will improve user satisfaction with content distribution services.

[1592] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[1593] Program processing steps

[1594] Step 1:

[1595] The user uses their smartphone's microphone to input voice data. The smartphone acquires this voice data and sends it to the server in real time.

[1596] Input: User's voice

[1597] Output: Audio data sent to the server

[1598] Step 2:

[1599] The server inputs the received audio data into a speech recognition API and converts it into text data. For example, Google Cloud Speech-to-Text is used as the speech recognition API.

[1600] Input: Audio data sent to the server

[1601] Output: Text data

[1602] Step 3:

[1603] The server inputs the converted text data into an emotion recognition engine to analyze the user's emotional state. IBM Watson Tone Analyzer is used as the emotion recognition engine.

[1604] Input: Text data

[1605] Output: User's emotional state (e.g., "frustration")

[1606] Step 4:

[1607] The server inputs the analyzed emotional state and text data into a generative AI to generate an appropriate response. OpenAI GPT-4 is used as the generative AI.

[1608] Input: Text data and user's emotional state

[1609] Output: Appropriate response

[1610] Step 5:

[1611] The server sends the generated response to the smartphone. The smartphone's user interface displays the received response.

[1612] Input: Generated response

[1613] Output: Response displayed on the smartphone's display screen

[1614] Specific operation of the process

[1615] Specific actions for Step 1:

[1616] When a user says to their smartphone, "The loading times are so slow, it's frustrating," the smartphone's microphone captures this audio and sends it to the server.

[1617] Specific actions in Step 2:

[1618] The server sends audio data to the Google Cloud Speech-to-Text API and receives this audio data as text. For example, the speech recognition API outputs the text "The loading time is so slow, it's frustrating."

[1619] Specific actions in Step 3:

[1620] The server sends the text message "The loading time is so slow, it's frustrating" to IBM Watson Tone Analyzer, and the analysis results in an emotional state of "frustration."

[1621] Specific actions in Step 4:

[1622] The server inputs the text data "The loading time is slow and frustrating" and the emotion state "frustrated" into OpenAI GPT-4 and generates a response such as "We apologize. We understand that you are experiencing inconvenience with the loading time. We will address this immediately."

[1623] Specific actions in Step 5:

[1624] The server sends the generated response to the smartphone. The smartphone's user interface then displays this response visually to the user.

[1625] In this way, a system can be realized that takes into account the user's emotions in real time and generates and displays appropriate responses.

[1626] 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.

[1627] 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.

[1628] 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 robot 414.

[1629] 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.

[1630] 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.

[1631] 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.

[1632] 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.

[1633] 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.

[1634] 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."

[1635] 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.

[1636] 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.

[1637] 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.

[1638] 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.

[1639] 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.

[1640] 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.

[1641] 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.

[1642] 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.

[1643] 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.

[1644] 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.

[1645] 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.

[1646] 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 as being incorporated by reference.

[1647] The following is further disclosed regarding the embodiments described above.

[1648] (Claim 1)

[1649] A means of converting acquired audio data into text in real time,

[1650] An artificial intelligence means that analyzes the converted text and generates an appropriate response,

[1651] A means for displaying the generated response on the call responder's display device,

[1652] A system that includes this.

[1653] (Claim 2)

[1654] The system according to claim 1, wherein the artificial intelligence means has previously learned industry-specific and department-specific knowledge.

[1655] (Claim 3)

[1656] The system according to claim 1, further comprising means for the call responder to adjust the response displayed on the display device.

[1657] "Example 1"

[1658] (Claim 1)

[1659] A means of converting acquired audio data into text in real time,

[1660] A means of analyzing the converted text and inputting it as a prompt sentence into a generative AI model that has been pre-trained with industry-specific and department-specific knowledge,

[1661] A means for displaying the response generated by the generation AI model on the call responder's display device,

[1662] A system that includes this.

[1663] (Claim 2)

[1664] The system according to claim 1, further comprising means for a call operator to adjust the response generated by the generation AI model.

[1665] (Claim 3)

[1666] The system according to claim 1, further comprising a user interface means that enables the call responder to refer to the response displayed on the display device in real time.

[1667] "Application Example 1"

[1668] (Claim 1)

[1669] A means of converting acquired audio data into text in real time,

[1670] An artificial intelligence means that analyzes the converted text and generates an appropriate response,

[1671] A means for displaying the generated response on a display device,

[1672] The display device is integrated into a wearable device, providing a means to provide information to the call responder in real time.

[1673] A system that includes this.

[1674] (Claim 2)

[1675] The system according to claim 1, wherein the artificial intelligence means has previously learned industry and departmental knowledge.

[1676] (Claim 3)

[1677] The system according to claim 1, further comprising means for the call responder to adjust the response displayed on the display device.

[1678] "Example 2 of combining an emotion engine"

[1679] (Claim 1)

[1680] A means of converting acquired audio data into text in real time,

[1681] An artificial intelligence means that analyzes the emotional state of converted text and voice and generates an appropriate response,

[1682] A means for displaying the generated response on the call responder's display device,

[1683] A system that includes this.

[1684] (Claim 2)

[1685] The system according to claim 1, wherein the artificial intelligence means pre-learns industry-specific and department-specific knowledge and further generates a response considering the emotional state of the voice.

[1686] (Claim 3)

[1687] The system according to claim 1, further comprising means for the call responder to adjust the response displayed on the display device.

[1688] "Application example 2 when combining with an emotional engine"

[1689] (Claim 1)

[1690] A means of converting acquired audio data into text in real time,

[1691] An artificial intelligence means that analyzes the converted text and generates an appropriate response,

[1692] An emotion engine means for analyzing the user's emotional state,

[1693] A means for displaying the response generated considering the emotion analysis results on the call responder's display device,

[1694] A system that includes this.

[1695] (Claim 2)

[1696] The system according to claim 1, wherein the artificial intelligence means has learned industry-specific and department-specific knowledge in advance, and further adjusts its response based on the sentiment analysis results.

[1697] (Claim 3)

[1698] The system according to claim 1, further comprising means for the call responder to adjust the response displayed on the display device. [Explanation of Symbols]

[1699] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of converting acquired audio data into text in real time, An artificial intelligence means that analyzes the converted text and generates an appropriate response, A means for displaying the generated response on the call responder's display device, A system that includes this.

2. The system according to claim 1, wherein the artificial intelligence means has previously learned industry-specific and department-specific knowledge.

3. The system according to claim 1, further comprising means for the call responder to adjust the response displayed on the display device.

Images