System

Abstract

A system is provided.SOLUTION: A system, comprising: means for collecting voice data of a user; means for converting the voice data into text; means for extracting features from the text data and the voice data; means for evaluating a skill level of the user based on the extracted features; means for presenting a customized practice scenario to the user; means for capturing and transmitting voice and facial expression data during practice of the user to a server; means for analyzing the captured data in real time and generating feedback to the user; and means for presenting the generated feedback to the user.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Previous vocational training for sales, presentations, and meetings required a large number of personnel for role-playing and customer interaction practice, making it difficult to achieve realistic interactions. Furthermore, new employees had few opportunities to learn how to behave effectively in meetings, and the quality of feedback varied. Providing feedback that includes elements that are difficult to verbalize, such as facial expressions and intonation, was difficult, and this presented a challenge. [Means for solving the problem]

[0005] The present invention solves the above-mentioned problems by providing a system including: means for collecting user voice data and converting it into text; means for extracting features from the text data and voice data; means for evaluating the user's skill level based on the extracted features; means for presenting a practice scenario that is individually tailored to the user; means for capturing the user's voice and facial expression data during practice and transmitting the captured data to a server; means for analyzing the captured data in real time and generating feedback for the user; and means for presenting the generated feedback to the user. This system enables realistic role-play and presentation practice without using any human resources, thereby reducing costs and improving learning opportunities.

[0006] "Voice data" refers to a digital voice file that records the user's speech.

[0007] "Text" is voice data converted into character information.

[0008] "Features" are numerical information such as speech stress, intonation, and speech rate extracted from speech data and text data.

[0009] The "skill level" is an index for evaluating the quality of a user's speech and facial expressions, and is an evaluation value calculated by a model.

[0010] A "practice scenario" is a hypothetical dialogue or presentation provided for a user to practice a particular skill.

[0011] "Capture" refers to the act or technique of collecting a user's voice and facial expressions in real time.

[0012] "Feedback" refers to areas for improvement and instructions generated based on the user's practice, including changes to voice volume, intonation, and facial expressions. [Brief explanation of the drawings]

[0013] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0014] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0015] First, the terms used in the following description will be explained.

[0016] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0017] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0018] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0019] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0020] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0021] [First embodiment]

[0022] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0023] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0024] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0025] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0026] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0027] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0028] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0029] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0030] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0031] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0032] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0033] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0034] The system and method of the present invention are an individually optimized dialogue simulator aimed at improving skills in sales, presentations, meetings, etc. This system collects and analyzes the user's voice data, provides individually tailored practice scenarios, and provides real-time feedback.

[0035] System program generation and processing details

[0036] 1. Data collection phase (server)

[0037] The server collects audio data and converts it into text. Audio recordings of specific sales proposals and presentations are collected from companies and stored as big data.

[0038] The collected voice data is converted into text data using automatic voice recognition technology.

[0039] 2. Data preprocessing phase (server)

[0040] The server extracts features from the text and audio data, such as speech stress, intonation patterns, and speaking rate.

[0041] 3. Model training phase (server)

[0042] Based on the extracted features, the server trains AI models to assess the user's skill level, including speech analysis models, intonation analysis models, and facial expression recognition models.

[0043] The trained model is saved and used to generate specialized training scenarios for the user.

[0044] 4. User authentication phase (terminal)

[0045] The user logs in to the system and enters authentication information, which the terminal sends to the server, where the user is identified through an authentication process.

[0046] 5. User Interaction Phase (Terminal)

[0047] The device retrieves a practice scenario specific to the user from the server and presents it to the user, who then begins practicing based on the scenario.

[0048] The device captures the user's voice and facial expression data while practicing and transmits it to the server in real time.

[0049] 6. Feedback Phase (Server and Device)

[0050] The server analyzes the received speech and facial expression data and generates real-time feedback to the user, including intonation corrections, improvements to specific expressions, and facial expression adjustments.

[0051] The device then presents the generated feedback to the user, providing specific advice such as, "It would be more effective if you raised the tone of your voice in this part," or "Smiling here will make a better impression."

[0052] Specific examples

[0053] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as "You should emphasize the intonation here" or "You might want to increase the tempo a little in this part." The generated feedback is provided to the user via the device, and the user uses it to carry out the next practice exercise.

[0054] In this way, the system of the present invention significantly improves the quality of vocational training and saves human resources by providing practice scenarios optimized for each user and enabling real-time feedback.In addition, by providing multifaceted feedback such as facial expressions and intonation, users can acquire practical and advanced skills.

[0055] The processing flow will be explained below.

[0056] Step 1:

[0057] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0058] Step 2:

[0059] The server converts the collected voice data into text using automatic speech recognition (ASR) technology, and the voice data is converted into text data and saved as text information.

[0060] Step 3:

[0061] The server extracts features from the text and audio data, calculating parameters such as voice stress, intonation patterns, and speaking rate.

[0062] Step 4:

[0063] The server trains AI models based on the preprocessed data to assess the user's skill level, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0064] Step 5:

[0065] Based on the results of the user's skill level assessment, the server generates specialized practice scenarios that provide specific content for the user to improve their skills.

[0066] Step 6:

[0067] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0068] Step 7:

[0069] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0070] Step 8:

[0071] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0072] Step 9:

[0073] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0074] Step 10:

[0075] The server generates specific feedback based on the analysis, such as "It would be more effective if you raised your voice tone in this part" or "If you smile briefly here, you'll make a better impression."

[0076] Step 11:

[0077] The device then presents the generated feedback to the user, who can use it to improve their next practice or presentation.

[0078] Step 12:

[0079] The user receives the provided feedback and practices the next time, adjusting the tone of their voice, intonation, and facial expressions based on the feedback.

[0080] The above is the specific flow of the program processing of the system aimed at improving user skills.

[0081] Example 1

[0082] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0083] Conventional vocational training systems have had the challenge of properly assessing individual users' skill levels and providing specialized practice scenarios and real-time feedback. Furthermore, it has been difficult to achieve both real-time performance and accuracy in the collection and analysis of voice and facial expression data. This has prevented users from improving their skills efficiently and effectively.

[0084] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0085] In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, means for extracting features from the text data and voice data, means for training a generative AI model that evaluates the user's skill level based on the extracted features, means for presenting an individually tailored practice scenario to the user, means for capturing the user's voice and facial expression data during practice and transmitting it to the server, means for analyzing the captured data in real time and generating feedback for the user, and means for presenting the generated feedback to the user. This makes it possible to accurately evaluate the user's skill level and provide individually optimized practice scenarios and real-time feedback.

[0086] "Voice data" refers to data that records in digital form the words spoken by a user.

[0087] "Text data" refers to data obtained by converting voice data into character information.

[0088] "Features" are important attributes or patterns that can be analyzed and extracted from speech or text data, such as speech stress, intonation patterns, and speaking rate.

[0089] "Skill level" is a standard for assessing a user's proficiency in a particular ability or technique.

[0090] A "generative AI model" is an artificial intelligence model that learns from training data based on features and evaluates and analyzes the user's skill level.

[0091] A "practice scenario" is a specific situation or task designed for users to practice, allowing them to practice in an environment that closely resembles a real-life work or presentation.

[0092] "Facial expression data" refers to data that captures a user's facial expressions and stores them in digital format.

[0093] "Feedback" refers to information that evaluates a user's performance during and after practice and provides suggestions and advice for improvement.

[0094] "Real-time analytics" refers to processing data and generating analytical results immediately as the data is collected.

[0095] "Capture" refers to recording the user's voice and facial expression data in digital format.

[0096] This invention is an individually optimized dialogue simulator aimed at improving skills in sales, presentations, meetings, etc. It collects and analyzes the user's voice data, provides individually specialized practice scenarios, and provides feedback in real time.

[0097] System configuration

[0098] The system is mainly composed of three components: a server, a terminal, and a user. The server collects data, analyzes it, trains models, and generates feedback. The terminal captures data and provides feedback. Users log in to the system, run practice scenarios, and receive feedback.

[0099] Hardware and software used

[0100] Server: Performs high-performance data analysis and model training using machine learning libraries such as TensorFlow and PyTorch, as well as the Google Cloud Speech-to-Text API.

[0101] Terminal: A device that captures voice and facial expressions and provides feedback to the user. The terminal software includes capture and real-time communication functions.

[0102] System Flow

[0103] 1. Collection and conversion of voice data (server)

[0104] The server collects audio data from sales proposals and presentations provided by companies and converts it into text data using the Google Cloud Speech-to-Text API. This process stores the audio as text information.

[0105] 2. Feature extraction (server)

[0106] The server extracts features such as voice intensity, intonation patterns, and speaking rate from the converted text and audio data using the Python librosa library.

[0107] 3. Training the AI ​​model (server)

[0108] Based on the extracted features, a generative AI model is trained using TensorFlow for speech analysis, intonation analysis, and facial expression recognition, and the model is saved after training.

[0109] 4. Presentation of practice scenario (terminal)

[0110] A user logs in to the system from a terminal and performs authentication. If authentication is successful, the server provides the user with a specialized practice scenario. The terminal displays this scenario, and the user begins practicing.

[0111] 5. Data capture and real-time transmission (terminal)

[0112] The device captures the user's voice and facial expression data while practicing and transmits it to the server in real time.

[0113] 6. Feedback Generation and Presentation (Server and Device)

[0114] The server analyzes the received data in real time and generates feedback, such as advice like "Raising your voice tone in this part will be more effective" or "Smiling here will make a better impression." The device then presents the generated feedback to the user.

[0115] Examples and prompts

[0116] For example, imagine a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as "You should emphasize the intonation here" or "It would be good to increase the tempo a little in this part." The generated feedback is provided to the user via the device, and the user uses this feedback to carry out the next practice session.

[0117] Example prompt sentence:

[0118] "Please give me feedback on where I should change my tone of voice when presenting a new product."

[0119] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0120] Step 1: Collecting voice data and converting it to text (server)

[0121] The server collects audio data of sales proposals and presentations provided by companies. The collected audio data is converted to text using the Google Cloud Speech-to-Text API. Specifically, the audio file is sent to the API, and the returned string is saved as a text file. The input is audio data, and the output is text data.

[0122] Step 2: Feature extraction (server)

[0123] The server extracts features from the converted text and audio data. Specifically, it uses the Python librosa library to analyze speech stress, intonation patterns, speaking rate, etc. The input is the text and audio data, and the output is the extracted feature data.

[0124] Step 3: Training the AI ​​model (server)

[0125] The server uses the extracted features to train a generative AI model. Specifically, it uses TensorFlow to create and train models for speech analysis, intonation analysis, and facial expression recognition. The input is the extracted feature data, and the output is the trained AI model.

[0126] Step 4: User authentication and access rights (device)

[0127] The user enters authentication information to log in to the system from their own terminal. The terminal sends this to the server, which executes the authentication process. The input is the user's authentication information, and the output is the authentication result (success or failure). If successful, access is granted, and if unsuccessful, an error message is returned.

[0128] Step 5: Acquire and present the practice scenario (terminal)

[0129] After successful authentication, the device retrieves a user-specific practice scenario from the server. This scenario is generated using a pre-trained AI model. The device displays the retrieved scenario to the user. The input is the successful user authentication status, and the output is the specialized practice scenario.

[0130] Step 6: Capture data during practice and send it in real time (on your device)

[0131] While the user practices based on the practice scenario, the device captures voice and facial expression data, which are then sent to the server in real time. The input is the user's voice and facial expression data, and the output is the captured data sent to the server.

[0132] Step 7: Generating and Presenting Feedback (Server and Device)

[0133] The server analyzes the received voice and facial expression data in real time and generates feedback. Specifically, it analyzes the user's intonation, speaking rate, facial expressions, etc. to identify areas for improvement. The generated feedback is sent to the device and presented to the user. The input is the captured data, and the output is the specific feedback content.

[0134] As a specific example of how it works, when a user gives a presentation, the server generates real-time feedback such as "It will be more effective if you raise the tone of your voice in this part" or "Smiling here will make a better impression," and provides it to the user via their device. Based on this feedback, the user can then practice the next part.

[0135] The above are the specific processing steps of the system.

[0136] (Application example 1)

[0137] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0138] There is a demand for a system that provides individually tailored practice scenarios based on the user's voice and facial expression data and provides feedback in real time. In customer service work, in particular, improving employee skills directly impacts sales and customer satisfaction, so effective and rapid training methods are essential. However, current systems face the challenge of making it difficult to provide detailed feedback in real time tailored to the user's individual skills and performance.

[0139] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0140] In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, and means for extracting features from the text data and voice data, thereby enabling the provision of practice scenarios based on individual user characteristics and real-time feedback.

[0141] "User voice data" refers to voice signals uttered by users, which are collected and analyzed to improve customer service skills.

[0142] "Means for converting to text" refers to technology that converts voice data into text information, and typically uses automatic speech recognition technology.

[0143] "Features" are patterns or attributes extracted from voice or text data and are used to evaluate a user's skills.

[0144] "Means for evaluating skill level" refers to technology that evaluates a user's ability and performance based on extracted features.

[0145] "Means for presenting practice scenarios" refers to technology that provides individually customized training content for the purpose of improving a user's skills.

[0146] "Voice and facial expression data" refers to data including the user's voice and facial expressions, and is used to analyze customer service performance.

[0147] "Means of real-time analysis" refers to technology that instantly processes and analyzes collected voice and facial expression data.

[0148] "Feedback" refers to improvements and advice provided to users based on the analysis results.

[0149] "Improving customer service skills" refers to techniques and methods aimed at improving a user's performance in customer service work.

[0150] This invention provides a training system for improving customer service skills. The system is configured by combining a smartphone terminal and a server, and collects user voice data and facial expression data, analyzing them in real time and providing feedback.

[0151] System Configuration

[0152] The system consists of the following main components:

[0153] 1. Data capture module (smartphone terminal)

[0154] Hardware: Smartphone camera, microphone.

[0155] Software: OpenCV (cv2), PyAudio. These are used to capture the user's voice and facial expression data in real time.

[0156] 2. Data transmission module (smartphone terminal)

[0157] Software: Requests library, used to send captured data to the server.

[0158] 3. Data analysis module (server)

[0159] Hardware: Cloud servers.

[0160] Software: TensorFlow, PyTorch, which are used to analyze the collected data and generate feedback.

[0161] 4. Feedback presentation module (smartphone device)

[0162] Software: A smartphone app that provides real-time feedback to users from the server.

[0163] Program processing

[0164] The smartphone device captures the user's voice and facial expressions and sends the data to a server. The captured data is acquired using the device's camera and microphone and processed by OpenCV and PyAudio. The data is then sent to the server using the Requests library.

[0165] The server analyzes the received voice and facial expression data using a generative AI model built with TensorFlow and PyTorch. This extracts features based on the user's intonation, vocal stress, speaking rate, and facial expressions, and evaluates the user's skill level.

[0166] Feedback is generated based on the analysis results from the server, providing specific advice such as "It will be more effective if you raise the tone of your voice in this part" or "Smiling here will make a better impression."

[0167] The smartphone device displays the feedback received from the server to the user in real time, allowing the user to immediately continue their training based on this feedback.

[0168] Examples and prompts

[0169] For example, when a sales clerk is explaining a new product to a customer, he or she places the smartphone in his or her breast pocket and launches the app. The app captures the clerk's voice and facial expressions and sends them to a server. The server analyzes the clerk's speech rate and facial expression patterns and provides real-time feedback such as, "You should smile more and speak more slowly when explaining the product's features."

[0170] Prompt Sentence Examples

[0171] "Please explain the features of our new product. Could you please give us some advice on what would be an effective explanation?"

[0172] "Please check the flow of conversations in real time while serving customers and let us know where we can improve."

[0173] "After you've finished explaining the product, please give us your feedback on what needs to be improved."

[0174] This is expected to enable efficient improvement of customer service skills and ultimately contribute to increased customer satisfaction.

[0175] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0176] Step 1: Data Capture

[0177] The smartphone device allows the user to perform a specific scenario (e.g., a new product presentation), and uses the device's camera and microphone to capture the user's voice and facial expression data in real time. This data is then sent raw to the next processing step.

[0178] Input: User's voice data, user's facial expression data

[0179] Output: Raw audio and facial expression data

[0180] Step 2: Send data

[0181] The captured voice and facial expression data is sent to the server using the smartphone's Requests library, where it is encoded and sent as an HTTP request.

[0182] Input: Raw audio and facial expression data

[0183] Output: The encoded data sent to the server.

[0184] Step 3: Convert audio data to text

[0185] The server converts the received voice data into text using automatic speech recognition (ASR), which converts the voice data into text data, making the voice content available as text.

[0186] Input: Encoded audio data

[0187] Output: Text data

[0188] Step 4: Feature extraction

[0189] The server extracts features from the text and audio data, such as speech stress, intonation, speaking rate, and text content, using speech analysis models and natural language processing (NLP) models.

[0190] Input: Text data, audio data

[0191] Output: Extracted features

[0192] Step 5: Assess your skill level

[0193] The server evaluates the user's skill level based on the extracted features, using a generative AI model to represent the user's customer service skills and presentation quality numerically and graphically.

[0194] Input: Extracted features

[0195] Output: User's skill level

[0196] Step 6: Provide a practice scenario

[0197] The server generates personalized practice scenarios based on the user's skill level, taking into account the user's weaknesses and areas for improvement, and suggests new scenarios.

[0198] Input: User's skill level

[0199] Output: A customized training scenario

[0200] Step 7: Capture user practice data

[0201] The user practices according to the provided practice scenario, while the smartphone again captures voice and facial expression data and sends them to the server. At the same time, the current practice status is monitored in real time.

[0202] Input: Voice data during practice, facial expression data during practice

[0203] Output: Practice data sent in real time

[0204] Step 8: Generate real-time feedback

[0205] The server analyzes the received practice data in real time and generates feedback for the user, including corrections to intonation, improvements to specific expressions, and adjustments to facial expressions.

[0206] Input: Voice data during practice, facial expression data during practice

[0207] Output: Real-time feedback

[0208] Step 9: Provide feedback

[0209] The smartphone device receives feedback from the server and presents it to the user in text, audio, visual, or other formats, allowing the user to use the feedback in their next practice.

[0210] Input: Real-time feedback

[0211] Output: Feedback presented to the user

[0212] This series of steps makes it possible to efficiently support users in improving their customer service skills.

[0213] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0214] This invention is a personalized dialogue simulator designed to improve skills in sales, presentations, meetings, and other areas. The system collects and analyzes the user's voice and facial expression data, provides customized practice scenarios, and provides real-time feedback. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it provides deeper feedback.

[0215] System program generation and processing details

[0216] 1. Data collection phase (server)

[0217] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0218] The collected voice data is converted into text using automatic speech recognition (ASR) technology and saved as text information.

[0219] 2. Data preprocessing phase (server)

[0220] The server extracts features from the text and audio data, including parameters such as voice stress, intonation patterns, and speaking rate.

[0221] 3. Model training phase (server)

[0222] The server trains AI models to assess the user's skill level based on the extracted features, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0223] The trained model is saved and used to generate specialized training scenarios for the user.

[0224] 4. Emotion engine integration (server)

[0225] The server incorporates a new emotion engine, which adds the ability to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation.

[0226] 5. User authentication phase (terminal)

[0227] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0228] 6. User Interaction Phase (Terminal)

[0229] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0230] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0231] 7. Feedback Phase (Server and Device)

[0232] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0233] The emotion engine recognizes the user's emotional state and assesses their stress level and motivation, which are then used to generate further feedback.

[0234] The server generates specific feedback based on the analysis, such as "You'll be more effective if you raise your voice tone here," "A quick smile here will make a better impression," or "You're under a lot of stress, so consider slowing down your practice."

[0235] The device displays the generated feedback to the user and provides specific instructions on how to improve for the next practice or presentation.

[0236] Specific examples

[0237] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the user's stress level and provides advice such as, "You may be too stressed and unable to concentrate on your practice. Take a break." The generated feedback is provided to the user via the device, and the user uses this information to carry out the next practice.

[0238] In this way, the system of the present invention significantly improves the quality of vocational training by providing practice scenarios optimized for each user and enabling real-time feedback. Furthermore, the introduction of an emotion engine provides detailed feedback according to the user's emotional state, enabling more practical and advanced skill improvement.

[0239] The processing flow will be explained below.

[0240] Step 1:

[0241] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0242] Step 2:

[0243] The server converts the collected voice data into text using automatic speech recognition (ASR) technology, and the voice data is converted into text data and saved as text information.

[0244] Step 3:

[0245] The server extracts features from the text and audio data, calculating parameters such as voice stress, intonation patterns, and speaking rate.

[0246] Step 4:

[0247] The server trains AI models based on the preprocessed data to assess the user's skill level, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0248] Step 5:

[0249] The server incorporates a new emotion engine, which adds the ability to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation.

[0250] Step 6:

[0251] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0252] Step 7:

[0253] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0254] Step 8:

[0255] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0256] Step 9:

[0257] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0258] Step 10:

[0259] The server generates specific feedback based on the analysis, such as "It would be more effective if you raised your voice tone in this part" or "If you smile briefly here, you'll make a better impression."

[0260] Step 11:

[0261] The emotion engine recognizes the user's emotional state from their voice and facial expression data and evaluates their stress level and motivation. For example, if the user's voice tone is low and their facial expression is stiff, it may determine that the user is feeling stressed.

[0262] Step 12:

[0263] The server generates further feedback based on the emotion engine's evaluation, such as advice like, "You may be under a lot of stress and not be able to concentrate on your practice. Take a break."

[0264] Step 13:

[0265] The device displays the generated feedback to the user and provides specific instructions on how to improve in the next practice or presentation. Based on the feedback, the user can adjust their voice tone, intonation, and facial expressions.

[0266] Step 14:

[0267] The user receives the provided feedback and practices the next time. Based on the feedback, the user can adjust the tone, intonation, and facial expressions of their voice to maximize the effectiveness of their training.

[0268] Example 2

[0269] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0270] Conventional vocational training systems have difficulty providing feedback that fully considers the characteristics and emotional state of individual users. This makes it difficult to efficiently improve users' practical skills. In addition, real-time feedback generation is often delayed, making it difficult to provide immediate improvements.

[0271] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, means for extracting features from the text data and voice data, means for evaluating the user's skill level based on the extracted features, means for recognizing the user's emotional state, means for presenting a practice scenario individually tailored to the user, means for capturing voice and facial expression data of the user during practice and transmitting the captured data to the server, means for analyzing the captured data in real time and generating feedback for the user, and means for presenting the generated feedback to the user. This makes it possible to provide a practice scenario individually optimized for the user and generate feedback instantly.

[0272] "Voice data" is digital data that records what the user has said.

[0273] "Text data" refers to character information generated from voice data using automatic voice recognition technology.

[0274] "Features" are parameters extracted from speech and facial expression data and used for analysis and model training. Specifically, they include speech stress, intonation patterns, speaking rate, etc.

[0275] The "means for assessing skill level" refers to an algorithm or model for assessing a user's ability or performance based on the extracted features.

[0276] The "means for recognizing emotional state" refers to an engine or technology that analyzes the user's voice data and facial expression data to grasp the user's emotional state, such as stress and motivation level.

[0277] The "means for presenting a practice scenario" is an interface or system for presenting practice content that is specialized for each individual user.

[0278] The "capturing means" refers to a device or software for collecting voice and facial expression data in real time while the user is practicing.

[0279] The "means for generating feedback" refers to an algorithm or system that analyzes the captured voice and facial expression data and presents areas for improvement to the user.

[0280] A "means for presenting feedback" is an interface or device that visually or audibly notifies the user of the generated feedback.

[0281] This invention is a personalized dialogue simulator designed to improve skills in sales, presentations, meetings, and other areas. The system collects and analyzes the user's voice and facial expression data, provides customized practice scenarios, and provides real-time feedback. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it provides deeper feedback.

[0282] First, the server collects recorded audio data of sales proposals and presentations provided by companies. This audio data is stored on the server as big data. The collected audio data is converted into text using automatic speech recognition (ASR) technology and saved as text information. A variety of automatic speech recognition APIs can be used for the ASR technology used.

[0283] The server then extracts features from the text and audio data, such as voice stress, intonation patterns, and speaking rate. A speech signal processing library is used to extract the features, and a computer vision library is used to analyze facial expression data.

[0284] The server then trains AI models based on these feature data. The trained models include a speech analysis model, an intonation analysis model, and a facial expression recognition model. The deep learning frameworks used include commonly used deep learning platforms.

[0285] The server also incorporates an emotion engine to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation. An emotion recognition API is available for this emotion analysis.

[0286] The user logs in to the system from their device and enters their authentication information. The device then sends this information to the server for user authentication. If authentication is successful, the device retrieves an individually tailored practice scenario from the server and presents it to the user. The user then begins practicing based on the displayed scenario. The device captures the user's voice and facial expression data in real time while they practice and sends the data to the server.

[0287] The server analyzes the received voice and facial expression data in real time. Analysis includes voice stress, changes in intonation, and facial expression recognition. The emotion engine recognizes the user's emotional state and evaluates their stress level and motivation. This evaluation result is used to generate further feedback. Based on the analysis results, the server generates specific feedback, such as "You should raise the tone of your voice in this part," "A quick smile here will create a good impression," or "You are under high stress, so consider slowing down your practice pace." The generated feedback is displayed to the user via their device, providing specific instructions on how to improve in their next practice or presentation.

[0288] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the user's stress level and provides advice such as, "You may be too stressed and unable to concentrate on your practice. Take a break." The generated feedback is provided to the user via the device, and the user uses this information to carry out the next practice.

[0289] Examples of input prompts include, "I would like the system to give me specific advice on what to pay attention to when presenting a new product." and "Please capture my voice and facial expression data during the presentation and evaluate it in real time."

[0290] In this way, the system of the present invention significantly improves the quality of vocational training by providing practice scenarios optimized for each user and generating immediate feedback. Furthermore, the introduction of an emotion engine provides detailed feedback according to the user's emotional state, enabling more practical and advanced skill improvement.

[0291] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0292] Step 1: Data collection

[0293] The server collects recorded audio data of sales proposals and presentations provided by companies. The server obtains audio data files from companies via HTTPS and stores them in data storage such as Google Cloud Storage. The input is the audio data file, and the output is the audio data file stored in the storage. The server converts the audio data into text data using the Google Cloud Speech-to-Text API and stores this text data in a database. The input is the audio data file, and the output is text data.

[0294] Step 2: Data Preprocessing

[0295] The server extracts features from text data and audio data. Features include voice stress, intonation patterns, speaking rate, etc. The Librosa library is used to extract features from the audio data and store them in a database. The input is audio data, and the output is feature data. The OpenCV library is used to analyze facial expression data, and facial expression features are also stored in the database. The input is facial expression image data, and the output is facial expression feature data.

[0296] Step 3: Model training

[0297] The server trains an AI model based on the collected feature data. Deep learning frameworks used include TensorFlow and PyTorch. For example, a speech analysis model is trained using TensorFlow and the model is saved. The input is the feature data, and the output is the trained AI model. Similarly, intonation analysis models and facial expression recognition models are trained and saved.

[0298] Step 4: Incorporating the Emotion Engine

[0299] The server incorporates an emotion engine into the system, adding the ability to recognize the user's emotional state. The emotion engine uses Affectiva or the Microsoft Emotion API. The API is used to analyze the user's emotions, and the results are stored in a database. The input is voice data and facial expression data, and the output is emotional state data.

[0300] Step 5: User authentication

[0301] A user logs in to the system from a terminal and enters authentication information. The terminal sends this information to the server, which then authenticates the user. OAuth2.0 or JWT is used for authentication. The input is the user authentication information, and the output is the authentication result.

[0302] Step 6: User Interaction

[0303] The device retrieves an individually tailored practice scenario from the server and presents it to the user. The user begins practicing based on the scenario displayed on the device. The device captures voice and facial expression data in real time and sends it to the server. The input is the practice scenario and the user's voice and facial expression data, and the output is the captured data.

[0304] Step 7: Generate feedback

[0305] The server analyzes the received voice and facial expression data in real time. The emotion engine also joins in to analyze the emotional state. For example, it analyzes the strength of the voice, changes in intonation, and recognizes facial expressions, and generates specific feedback. Based on the analysis results, it generates feedback such as "You should raise the tone of your voice in this part" or "If you smile briefly here, it will make a good impression." The input is the captured voice and facial expression data, and the output is the feedback data. The generated feedback is displayed to the user via the terminal.

[0306] The above is the specific processing flow of the program of this system.

[0307] (Application example 2)

[0308] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0309] Improving skills in sales, customer service, and presentations requires a lot of time and effort through manual education and training, but it is difficult to provide individually optimized feedback. There is also a lack of a system for evaluating the effectiveness of training in real time and immediately proposing improvement measures. Furthermore, it is difficult to maintain motivation for practice because feedback cannot take into account the emotional state of the trainee. This creates the problem of insufficient progress in improving customer service and presentation skills.

[0310] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice information, means for converting the voice information into character data, means for extracting features from the character data and the voice information, means for evaluating the user's skill level based on the extracted features, means for presenting an individually tailored practice scenario to the user, means for capturing the user's voice and facial expression information during practice and transmitting it to the server, means for analyzing the captured information in real time and generating feedback for the user, means for recognizing the user's emotional state, and means for presenting the generated feedback to the user. This makes it possible to effectively improve the user's skills through individually optimized practice scenarios and real-time feedback, and furthermore, by providing detailed feedback according to the user's emotional state, it is possible to achieve training effectiveness and maintain motivation.

[0311] "Voice information" refers to digitally recorded audio data, including the user's speech, tone of voice, intonation, and the like.

[0312] "Text data" refers to data that has been converted into text format by analyzing voice information, and is a written representation of what the user has said.

[0313] "Features" refer to measurable characteristics of a user's speech, such as stress, intonation, and speech rate, extracted from audio information and text data.

[0314] The "skill level" is an index that indicates the technical proficiency of the user's speaking style and facial expressions, evaluated based on the extracted feature amounts.

[0315] A "practice scenario" is an individually specialized practice scenario provided by the system for the purpose of improving the user's skills, and its content is customized according to the user's skill level.

[0316] "Facial expression information" refers to data that digitally records a user's facial expressions, including facial movements and emotional expressions.

[0317] The "emotional state" refers to the user's emotional state, which is recognized based on the user's facial expression information and voice information, and includes, for example, stress level and motivation level.

[0318] "Feedback" is specific advice about areas for improvement and effective practice methods for the user, generated based on analyzed voice information, text data, facial expression information, and emotional state.

[0319] "Real-time analysis" means processing the user's voice information and facial expression information instantly and obtaining analysis results instantly.

[0320] The present invention is an individually optimized dialogue simulator aimed at improving customer service skills in brick-and-mortar stores. This system involves communication between a server and a terminal, and provides users with optimized practice scenarios and feedback. A specific embodiment of this system is described below.

[0321] Data collection phase (server)

[0322] The server collects user voice information, which is uttered when the user practices customer service skills in a physical store and is recorded in real time using a smartphone or smart glasses.

[0323] Data preprocessing phase (server)

[0324] The collected speech information is converted into text data using automatic speech recognition (ASR) technology, such as Google Cloud Speech to Text. The Librosa library is then used to extract features from the speech data, including speech stress, intonation patterns, and speaking rate.

[0325] Model training phase (server)

[0326] The server trains AI models to assess the user's skill level based on the extracted features. These models include speech analysis, intonation analysis, and facial expression recognition. The trained models are saved and used to generate personalized practice scenarios for the user.

[0327] Emotion engine integration (server)

[0328] The server incorporates an emotion engine that recognizes the user's emotional state and analyzes it. The emotion engine is built using a deep learning library such as TensorFlow and evaluates the user's emotional state based on their voice and facial expression information.

[0329] User authentication phase (terminal)

[0330] The user logs in to the system from a terminal and begins practicing. The terminal can be a smartphone or smart glasses, and has the ability to capture the user's voice and facial expression information.

[0331] User Interaction Phase (Terminal)

[0332] The device retrieves an individually tailored practice scenario from the server and presents it to the user. The user begins practicing based on the displayed scenario, capturing voice and facial expression information in real time during the practice process and sending it to the server.

[0333] Feedback Phase (Server and Device)

[0334] The server analyzes the captured voice and facial expression information in real time and generates feedback. This feedback includes suggestions for improving voice stress, intonation, and facial expressions, as well as advice based on the user's emotional state. For example, the server provides specific feedback to the user, such as "It would be more effective if you raised the tone of your voice in this part," or "If you smile briefly here, you will make a good impression." The generated feedback is presented to the user via their device and can be used for future practice or when serving customers in physical stores.

[0335] Specific examples

[0336] For example, imagine a staff member introducing a new product. The staff member logs in to the system and starts a specialized practice scenario. The terminal captures the voice and facial expression information as the staff member explains the product's features and sends it to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the staff member's stress level and provides advice such as, "You may be too stressed and not able to concentrate on your practice. Take a break." The generated feedback is provided to the staff member via the terminal, and they use it to carry out their next practice.

[0337] Prompt Sentence Examples

[0338] Please generate a new customer service scenario. This scenario should be about a female customer in her twenties asking about a new product. The scenario should include the conversation with the customer, the information provided by the staff, and the features of the recommended product.

[0339] As described above, the system of the present invention supports the improvement of customer service skills in brick-and-mortar stores by providing practice scenarios optimized for each user and enabling real-time feedback. Furthermore, the introduction of an emotion engine allows for more detailed feedback based on the user's emotional state, enabling more effective training.

[0340] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0341] Step 1:

[0342] The server collects the user's voice information. This voice information is spoken by the user while practicing customer service skills and is recorded in real time using a device (smartphone or smart glasses). The input is the user's voice, and the output is digital voice data.

[0343] Step 2:

[0344] The server converts the collected voice information into text data using automatic speech recognition (ASR) technology. At this stage, the input is voice data and the output is text data. Specifically, an ASR service such as Google Cloud Speech to Text is used.

[0345] Step 3:

[0346] The server extracts features from text data and speech information. It uses the Librosa library to calculate features such as speech stress, intonation patterns, and speaking rate. The input is speech data and text data, and the output is a set of features.

[0347] Step 4:

[0348] The server evaluates the user's skill level based on the extracted features. This evaluation uses pre-trained AI models (voice analysis model, intonation analysis model, and facial expression recognition model). The input is the features, and the output is the skill level evaluation result.

[0349] Step 5:

[0350] The server generates a practice scenario specifically tailored for each user and sends it to the device. The scenario is customized based on the user's skill level assessment results. The input is the skill level assessment results, and the output is the practice scenario. The generative AI model generates the scenario according to the prompt sentence.

[0351] Step 6:

[0352] The terminal presents a practice scenario to the user, who then practices according to the scenario. The input is the practice scenario, and the output is the content of the practice performed by the user. Specifically, the scenario is displayed on the screen or guided by voice.

[0353] Step 7:

[0354] The device captures the user's voice and facial expression information while they practice and sends it to the server in real time. The input is the user's voice and facial expression information, and the output is the captured data. Specifically, the device's camera and microphone are always active.

[0355] Step 8:

[0356] The server analyzes the captured speech and facial expression information in real time and generates feedback for the user, including suggestions for improving intonation, vocal stress, facial expression, and emotional state. The input is the captured speech and facial expression information, and the output is the feedback.

[0357] Step 9:

[0358] The server uses an emotion engine to recognize the user's emotional state and reflects that information in the feedback. The input is captured voice and facial expression information, and the output is the evaluation result of the emotional state. Specifically, emotion analysis is performed using TensorFlow.

[0359] Step 10:

[0360] The terminal presents the generated feedback to the user. The input is the feedback, and the output is improvements or advice that is displayed or announced to the user by voice. Specific operations include displaying the feedback as text on the screen or announcing it by voice.

[0361] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0362] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0363] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0364] [Second embodiment]

[0365] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0366] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0367] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0368] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0369] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0370] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0371] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0372] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0373] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0374] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0375] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0376] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0377] The system and method of the present invention are an individually optimized dialogue simulator aimed at improving skills in sales, presentations, meetings, etc. This system collects and analyzes the user's voice data, provides individually tailored practice scenarios, and provides real-time feedback.

[0378] System program generation and processing details

[0379] 1. Data collection phase (server)

[0380] The server collects audio data and converts it into text. Audio recordings of specific sales proposals and presentations are collected from companies and stored as big data.

[0381] The collected voice data is converted into text data using automatic voice recognition technology.

[0382] 2. Data preprocessing phase (server)

[0383] The server extracts features from the text and audio data, such as speech stress, intonation patterns, and speaking rate.

[0384] 3. Model training phase (server)

[0385] Based on the extracted features, the server trains AI models to assess the user's skill level, including speech analysis models, intonation analysis models, and facial expression recognition models.

[0386] The trained model is saved and used to generate specialized training scenarios for the user.

[0387] 4. User authentication phase (terminal)

[0388] The user logs in to the system and enters authentication information, which the terminal sends to the server, where the user is identified through an authentication process.

[0389] 5. User Interaction Phase (Terminal)

[0390] The device retrieves a practice scenario specific to the user from the server and presents it to the user, who then begins practicing based on the scenario.

[0391] The device captures the user's voice and facial expression data while practicing and transmits it to the server in real time.

[0392] 6. Feedback Phase (Server and Device)

[0393] The server analyzes the received speech and facial expression data and generates real-time feedback to the user, including intonation corrections, improvements to specific expressions, and facial expression adjustments.

[0394] The device then presents the generated feedback to the user, providing specific advice such as, "It would be more effective if you raised the tone of your voice in this part," or "Smiling here will make a better impression."

[0395] Specific examples

[0396] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as "You should emphasize the intonation here" or "You might want to increase the tempo a little in this part." The generated feedback is provided to the user via the device, and the user uses it to carry out the next practice exercise.

[0397] In this way, the system of the present invention significantly improves the quality of vocational training and saves human resources by providing practice scenarios optimized for each user and enabling real-time feedback.In addition, by providing multifaceted feedback such as facial expressions and intonation, users can acquire practical and advanced skills.

[0398] The processing flow will be explained below.

[0399] Step 1:

[0400] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0401] Step 2:

[0402] The server converts the collected voice data into text using automatic speech recognition (ASR) technology, and the voice data is converted into text data and saved as text information.

[0403] Step 3:

[0404] The server extracts features from the text and audio data, calculating parameters such as voice stress, intonation patterns, and speaking rate.

[0405] Step 4:

[0406] The server trains AI models based on the preprocessed data to assess the user's skill level, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0407] Step 5:

[0408] Based on the results of the user's skill level assessment, the server generates specialized practice scenarios that provide specific content for the user to improve their skills.

[0409] Step 6:

[0410] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0411] Step 7:

[0412] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0413] Step 8:

[0414] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0415] Step 9:

[0416] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0417] Step 10:

[0418] The server generates specific feedback based on the analysis, such as "It would be more effective if you raised your voice tone in this part" or "If you smile briefly here, you'll make a better impression."

[0419] Step 11:

[0420] The device then presents the generated feedback to the user, who can use it to improve their next practice or presentation.

[0421] Step 12:

[0422] The user receives the provided feedback and practices the next time, adjusting the tone of their voice, intonation, and facial expressions based on the feedback.

[0423] The above is the specific flow of the program processing of the system aimed at improving user skills.

[0424] Example 1

[0425] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0426] Conventional vocational training systems have had the challenge of properly assessing individual users' skill levels and providing specialized practice scenarios and real-time feedback. Furthermore, it has been difficult to achieve both real-time performance and accuracy in the collection and analysis of voice and facial expression data. This has prevented users from improving their skills efficiently and effectively.

[0427] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0428] In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, means for extracting features from the text data and voice data, means for training a generative AI model that evaluates the user's skill level based on the extracted features, means for presenting an individually tailored practice scenario to the user, means for capturing the user's voice and facial expression data during practice and transmitting it to the server, means for analyzing the captured data in real time and generating feedback for the user, and means for presenting the generated feedback to the user. This makes it possible to accurately evaluate the user's skill level and provide individually optimized practice scenarios and real-time feedback.

[0429] "Voice data" refers to data that records in digital form the words spoken by a user.

[0430] "Text data" refers to data obtained by converting voice data into character information.

[0431] "Features" are important attributes or patterns that can be analyzed and extracted from speech or text data, such as speech stress, intonation patterns, and speaking rate.

[0432] "Skill level" is a standard for assessing a user's proficiency in a particular ability or technique.

[0433] A "generative AI model" is an artificial intelligence model that learns from training data based on features and evaluates and analyzes the user's skill level.

[0434] A "practice scenario" is a specific situation or task designed for users to practice, allowing them to practice in an environment that closely resembles a real-life work or presentation.

[0435] "Facial expression data" refers to data that captures a user's facial expressions and stores them in digital format.

[0436] "Feedback" refers to information that evaluates a user's performance during and after practice and provides suggestions and advice for improvement.

[0437] "Real-time analytics" refers to processing data and generating analytical results immediately as the data is collected.

[0438] "Capture" refers to recording the user's voice and facial expression data in digital format.

[0439] This invention is an individually optimized dialogue simulator aimed at improving skills in sales, presentations, meetings, etc. It collects and analyzes the user's voice data, provides individually specialized practice scenarios, and provides feedback in real time.

[0440] System configuration

[0441] The system is mainly composed of three components: a server, a terminal, and a user. The server collects data, analyzes it, trains models, and generates feedback. The terminal captures data and provides feedback. Users log in to the system, run practice scenarios, and receive feedback.

[0442] Hardware and software used

[0443] Server: Performs high-performance data analysis and model training using machine learning libraries such as TensorFlow and PyTorch, as well as the Google Cloud Speech-to-Text API.

[0444] Terminal: A device that captures voice and facial expressions and provides feedback to the user. The terminal software includes capture and real-time communication functions.

[0445] System Flow

[0446] 1. Collection and conversion of voice data (server)

[0447] The server collects audio data from sales proposals and presentations provided by companies and converts it into text data using the Google Cloud Speech-to-Text API. This process stores the audio as text information.

[0448] 2. Feature extraction (server)

[0449] The server extracts features such as voice intensity, intonation patterns, and speaking rate from the converted text and audio data using the Python librosa library.

[0450] 3. Training the AI ​​model (server)

[0451] Based on the extracted features, a generative AI model is trained using TensorFlow for speech analysis, intonation analysis, and facial expression recognition, and the model is saved after training.

[0452] 4. Presentation of practice scenario (terminal)

[0453] A user logs in to the system from a terminal and performs authentication. If authentication is successful, the server provides the user with a specialized practice scenario. The terminal displays this scenario, and the user begins practicing.

[0454] 5. Data capture and real-time transmission (terminal)

[0455] The device captures the user's voice and facial expression data while practicing and transmits it to the server in real time.

[0456] 6. Feedback Generation and Presentation (Server and Device)

[0457] The server analyzes the received data in real time and generates feedback, such as advice like "Raising your voice tone in this part will be more effective" or "Smiling here will make a better impression." The device then presents the generated feedback to the user.

[0458] Examples and prompts

[0459] For example, imagine a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as "You should emphasize the intonation here" or "It would be good to increase the tempo a little in this part." The generated feedback is provided to the user via the device, and the user uses this feedback to carry out the next practice session.

[0460] Example prompt sentence:

[0461] "Please give me feedback on where I should change my tone of voice when presenting a new product."

[0462] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0463] Step 1: Collecting voice data and converting it to text (server)

[0464] The server collects audio data of sales proposals and presentations provided by companies. The collected audio data is converted to text using the Google Cloud Speech-to-Text API. Specifically, the audio file is sent to the API, and the returned string is saved as a text file. The input is audio data, and the output is text data.

[0465] Step 2: Feature extraction (server)

[0466] The server extracts features from the converted text and audio data. Specifically, it uses the Python librosa library to analyze speech stress, intonation patterns, speaking rate, etc. The input is the text and audio data, and the output is the extracted feature data.

[0467] Step 3: Training the AI ​​model (server)

[0468] The server uses the extracted features to train a generative AI model. Specifically, it uses TensorFlow to create and train models for speech analysis, intonation analysis, and facial expression recognition. The input is the extracted feature data, and the output is the trained AI model.

[0469] Step 4: User authentication and access rights (device)

[0470] The user enters authentication information to log in to the system from their own terminal. The terminal sends this to the server, which executes the authentication process. The input is the user's authentication information, and the output is the authentication result (success or failure). If successful, access is granted, and if unsuccessful, an error message is returned.

[0471] Step 5: Acquire and present the practice scenario (terminal)

[0472] After successful authentication, the device retrieves a user-specific practice scenario from the server. This scenario is generated using a pre-trained AI model. The device displays the retrieved scenario to the user. The input is the successful user authentication status, and the output is the specialized practice scenario.

[0473] Step 6: Capture data during practice and send it in real time (on your device)

[0474] While the user practices based on the practice scenario, the device captures voice and facial expression data, which are then sent to the server in real time. The input is the user's voice and facial expression data, and the output is the captured data sent to the server.

[0475] Step 7: Generating and Presenting Feedback (Server and Device)

[0476] The server analyzes the received voice and facial expression data in real time and generates feedback. Specifically, it analyzes the user's intonation, speaking rate, facial expressions, etc. to identify areas for improvement. The generated feedback is sent to the device and presented to the user. The input is the captured data, and the output is the specific feedback content.

[0477] As a specific example of how it works, when a user gives a presentation, the server generates real-time feedback such as "It will be more effective if you raise the tone of your voice in this part" or "Smiling here will make a better impression," and provides it to the user via their device. Based on this feedback, the user can then practice the next part.

[0478] The above are the specific processing steps of the system.

[0479] (Application example 1)

[0480] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0481] There is a demand for a system that provides individually tailored practice scenarios based on the user's voice and facial expression data and provides feedback in real time. In customer service work, in particular, improving employee skills directly impacts sales and customer satisfaction, so effective and rapid training methods are essential. However, current systems face the challenge of making it difficult to provide detailed feedback in real time tailored to the user's individual skills and performance.

[0482] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0483] In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, and means for extracting features from the text data and voice data, thereby enabling the provision of practice scenarios based on individual user characteristics and real-time feedback.

[0484] "User voice data" refers to voice signals uttered by users, which are collected and analyzed to improve customer service skills.

[0485] "Means for converting to text" refers to technology that converts voice data into text information, and typically uses automatic speech recognition technology.

[0486] "Features" are patterns or attributes extracted from voice or text data and are used to evaluate a user's skills.

[0487] "Means for evaluating skill level" refers to technology that evaluates a user's ability and performance based on extracted features.

[0488] "Means for presenting practice scenarios" refers to technology that provides individually customized training content for the purpose of improving a user's skills.

[0489] "Voice and facial expression data" refers to data including the user's voice and facial expressions, and is used to analyze customer service performance.

[0490] "Means of real-time analysis" refers to technology that instantly processes and analyzes collected voice and facial expression data.

[0491] "Feedback" refers to improvements and advice provided to users based on the analysis results.

[0492] "Improving customer service skills" refers to techniques and methods aimed at improving a user's performance in customer service work.

[0493] This invention provides a training system for improving customer service skills. The system is configured by combining a smartphone terminal and a server, and collects user voice data and facial expression data, analyzing them in real time and providing feedback.

[0494] System Configuration

[0495] The system consists of the following main components:

[0496] 1. Data capture module (smartphone terminal)

[0497] Hardware: Smartphone camera, microphone.

[0498] Software: OpenCV (cv2), PyAudio. These are used to capture the user's voice and facial expression data in real time.

[0499] 2. Data transmission module (smartphone terminal)

[0500] Software: Requests library, used to send captured data to the server.

[0501] 3. Data analysis module (server)

[0502] Hardware: Cloud servers.

[0503] Software: TensorFlow, PyTorch, which are used to analyze the collected data and generate feedback.

[0504] 4. Feedback presentation module (smartphone device)

[0505] Software: A smartphone app that provides real-time feedback to users from the server.

[0506] Program processing

[0507] The smartphone device captures the user's voice and facial expressions and sends the data to a server. The captured data is acquired using the device's camera and microphone and processed by OpenCV and PyAudio. The data is then sent to the server using the Requests library.

[0508] The server analyzes the received voice and facial expression data using a generative AI model built with TensorFlow and PyTorch. This extracts features based on the user's intonation, vocal stress, speaking rate, and facial expressions, and evaluates the user's skill level.

[0509] Feedback is generated based on the analysis results from the server, providing specific advice such as "It will be more effective if you raise the tone of your voice in this part" or "Smiling here will make a better impression."

[0510] The smartphone device displays the feedback received from the server to the user in real time, allowing the user to immediately continue their training based on this feedback.

[0511] Examples and prompts

[0512] For example, when a sales clerk is explaining a new product to a customer, he or she places the smartphone in his or her breast pocket and launches the app. The app captures the clerk's voice and facial expressions and sends them to a server. The server analyzes the clerk's speech rate and facial expression patterns and provides real-time feedback such as, "You should smile more and speak more slowly when explaining the product's features."

[0513] Prompt Sentence Examples

[0514] "Please explain the features of our new product. Could you please give us some advice on what would be an effective explanation?"

[0515] "Please check the flow of conversations in real time while serving customers and let us know where we can improve."

[0516] "After you've finished explaining the product, please give us your feedback on what needs to be improved."

[0517] This is expected to enable efficient improvement of customer service skills and ultimately contribute to increased customer satisfaction.

[0518] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0519] Step 1: Data Capture

[0520] The smartphone device allows the user to perform a specific scenario (e.g., a new product presentation), and uses the device's camera and microphone to capture the user's voice and facial expression data in real time. This data is then sent raw to the next processing step.

[0521] Input: User's voice data, user's facial expression data

[0522] Output: Raw audio and facial expression data

[0523] Step 2: Send data

[0524] The captured voice and facial expression data is sent to the server using the smartphone's Requests library, where it is encoded and sent as an HTTP request.

[0525] Input: Raw audio and facial expression data

[0526] Output: The encoded data sent to the server.

[0527] Step 3: Convert audio data to text

[0528] The server converts the received voice data into text using automatic speech recognition (ASR), which converts the voice data into text data, making the voice content available as text.

[0529] Input: Encoded audio data

[0530] Output: Text data

[0531] Step 4: Feature extraction

[0532] The server extracts features from the text and audio data, such as speech stress, intonation, speaking rate, and text content, using speech analysis models and natural language processing (NLP) models.

[0533] Input: Text data, audio data

[0534] Output: Extracted features

[0535] Step 5: Assess your skill level

[0536] The server evaluates the user's skill level based on the extracted features, using a generative AI model to represent the user's customer service skills and presentation quality numerically and graphically.

[0537] Input: Extracted features

[0538] Output: User's skill level

[0539] Step 6: Provide a practice scenario

[0540] The server generates personalized practice scenarios based on the user's skill level, taking into account the user's weaknesses and areas for improvement, and suggests new scenarios.

[0541] Input: User's skill level

[0542] Output: A customized training scenario

[0543] Step 7: Capture user practice data

[0544] The user practices according to the provided practice scenario, while the smartphone again captures voice and facial expression data and sends them to the server. At the same time, the current practice status is monitored in real time.

[0545] Input: Voice data during practice, facial expression data during practice

[0546] Output: Practice data sent in real time

[0547] Step 8: Generate real-time feedback

[0548] The server analyzes the received practice data in real time and generates feedback for the user, including corrections to intonation, improvements to specific expressions, and adjustments to facial expressions.

[0549] Input: Voice data during practice, facial expression data during practice

[0550] Output: Real-time feedback

[0551] Step 9: Provide feedback

[0552] The smartphone device receives feedback from the server and presents it to the user in text, audio, visual, or other formats, allowing the user to use the feedback in their next practice.

[0553] Input: Real-time feedback

[0554] Output: Feedback presented to the user

[0555] This series of steps makes it possible to efficiently support users in improving their customer service skills.

[0556] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0557] This invention is a personalized dialogue simulator designed to improve skills in sales, presentations, meetings, and other areas. The system collects and analyzes the user's voice and facial expression data, provides customized practice scenarios, and provides real-time feedback. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it provides deeper feedback.

[0558] System program generation and processing details

[0559] 1. Data collection phase (server)

[0560] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0561] The collected voice data is converted into text using automatic speech recognition (ASR) technology and saved as text information.

[0562] 2. Data preprocessing phase (server)

[0563] The server extracts features from the text and audio data, including parameters such as voice stress, intonation patterns, and speaking rate.

[0564] 3. Model training phase (server)

[0565] The server trains AI models to assess the user's skill level based on the extracted features, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0566] The trained model is saved and used to generate specialized training scenarios for the user.

[0567] 4. Emotion engine integration (server)

[0568] The server incorporates a new emotion engine, which adds the ability to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation.

[0569] 5. User authentication phase (terminal)

[0570] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0571] 6. User Interaction Phase (Terminal)

[0572] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0573] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0574] 7. Feedback Phase (Server and Device)

[0575] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0576] The emotion engine recognizes the user's emotional state and assesses their stress level and motivation, which are then used to generate further feedback.

[0577] The server generates specific feedback based on the analysis, such as "You'll be more effective if you raise your voice tone here," "A quick smile here will make a better impression," or "You're under a lot of stress, so consider slowing down your practice."

[0578] The device displays the generated feedback to the user and provides specific instructions on how to improve for the next practice or presentation.

[0579] Specific examples

[0580] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the user's stress level and provides advice such as, "You may be too stressed and unable to concentrate on your practice. Take a break." The generated feedback is provided to the user via the device, and the user uses this information to carry out the next practice.

[0581] In this way, the system of the present invention significantly improves the quality of vocational training by providing practice scenarios optimized for each user and enabling real-time feedback. Furthermore, the introduction of an emotion engine provides detailed feedback according to the user's emotional state, enabling more practical and advanced skill improvement.

[0582] The processing flow will be explained below.

[0583] Step 1:

[0584] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0585] Step 2:

[0586] The server converts the collected voice data into text using automatic speech recognition (ASR) technology, and the voice data is converted into text data and saved as text information.

[0587] Step 3:

[0588] The server extracts features from the text and audio data, calculating parameters such as voice stress, intonation patterns, and speaking rate.

[0589] Step 4:

[0590] The server trains AI models based on the preprocessed data to assess the user's skill level, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0591] Step 5:

[0592] The server incorporates a new emotion engine, which adds the ability to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation.

[0593] Step 6:

[0594] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0595] Step 7:

[0596] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0597] Step 8:

[0598] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0599] Step 9:

[0600] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0601] Step 10:

[0602] The server generates specific feedback based on the analysis, such as "It would be more effective if you raised your voice tone in this part" or "If you smile briefly here, you'll make a better impression."

[0603] Step 11:

[0604] The emotion engine recognizes the user's emotional state from their voice and facial expression data and evaluates their stress level and motivation. For example, if the user's voice tone is low and their facial expression is stiff, it may determine that the user is feeling stressed.

[0605] Step 12:

[0606] The server generates further feedback based on the emotion engine's evaluation, such as advice like, "You may be under a lot of stress and not be able to concentrate on your practice. Take a break."

[0607] Step 13:

[0608] The device displays the generated feedback to the user and provides specific instructions on how to improve in the next practice or presentation. Based on the feedback, the user can adjust their voice tone, intonation, and facial expressions.

[0609] Step 14:

[0610] The user receives the provided feedback and practices the next time. Based on the feedback, the user can adjust the tone, intonation, and facial expressions of their voice to maximize the effectiveness of their training.

[0611] Example 2

[0612] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0613] Conventional vocational training systems have difficulty providing feedback that fully considers the characteristics and emotional state of individual users. This makes it difficult to efficiently improve users' practical skills. In addition, real-time feedback generation is often delayed, making it difficult to provide immediate improvements.

[0614] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, means for extracting features from the text data and voice data, means for evaluating the user's skill level based on the extracted features, means for recognizing the user's emotional state, means for presenting a practice scenario individually tailored to the user, means for capturing voice and facial expression data of the user during practice and transmitting the captured data to the server, means for analyzing the captured data in real time and generating feedback for the user, and means for presenting the generated feedback to the user. This makes it possible to provide a practice scenario individually optimized for the user and generate feedback instantly.

[0615] "Voice data" is digital data that records what the user has said.

[0616] "Text data" refers to character information generated from voice data using automatic voice recognition technology.

[0617] "Features" are parameters extracted from speech and facial expression data and used for analysis and model training. Specifically, they include speech stress, intonation patterns, speaking rate, etc.

[0618] The "means for assessing skill level" refers to an algorithm or model for assessing a user's ability or performance based on the extracted features.

[0619] The "means for recognizing emotional state" refers to an engine or technology that analyzes the user's voice data and facial expression data to grasp the user's emotional state, such as stress and motivation level.

[0620] The "means for presenting a practice scenario" is an interface or system for presenting practice content that is specialized for each individual user.

[0621] The "capturing means" refers to a device or software for collecting voice and facial expression data in real time while the user is practicing.

[0622] The "means for generating feedback" refers to an algorithm or system that analyzes the captured voice and facial expression data and presents areas for improvement to the user.

[0623] A "means for presenting feedback" is an interface or device that visually or audibly notifies the user of the generated feedback.

[0624] This invention is a personalized dialogue simulator designed to improve skills in sales, presentations, meetings, and other areas. The system collects and analyzes the user's voice and facial expression data, provides customized practice scenarios, and provides real-time feedback. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it provides deeper feedback.

[0625] First, the server collects recorded audio data of sales proposals and presentations provided by companies. This audio data is stored on the server as big data. The collected audio data is converted into text using automatic speech recognition (ASR) technology and saved as text information. A variety of automatic speech recognition APIs can be used for the ASR technology used.

[0626] The server then extracts features from the text and audio data, such as voice stress, intonation patterns, and speaking rate. A speech signal processing library is used to extract the features, and a computer vision library is used to analyze facial expression data.

[0627] The server then trains AI models based on these feature data. The trained models include a speech analysis model, an intonation analysis model, and a facial expression recognition model. The deep learning frameworks used include commonly used deep learning platforms.

[0628] The server also incorporates an emotion engine to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation. An emotion recognition API is available for this emotion analysis.

[0629] The user logs in to the system from their device and enters their authentication information. The device then sends this information to the server for user authentication. If authentication is successful, the device retrieves an individually tailored practice scenario from the server and presents it to the user. The user then begins practicing based on the displayed scenario. The device captures the user's voice and facial expression data in real time while they practice and sends the data to the server.

[0630] The server analyzes the received voice and facial expression data in real time. Analysis includes voice stress, changes in intonation, and facial expression recognition. The emotion engine recognizes the user's emotional state and evaluates their stress level and motivation. This evaluation result is used to generate further feedback. Based on the analysis results, the server generates specific feedback, such as "You should raise the tone of your voice in this part," "A quick smile here will create a good impression," or "You are under high stress, so consider slowing down your practice pace." The generated feedback is displayed to the user via their device, providing specific instructions on how to improve in their next practice or presentation.

[0631] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the user's stress level and provides advice such as, "You may be too stressed and unable to concentrate on your practice. Take a break." The generated feedback is provided to the user via the device, and the user uses this information to carry out the next practice.

[0632] Examples of input prompts include, "I would like the system to give me specific advice on what to pay attention to when presenting a new product." and "Please capture my voice and facial expression data during the presentation and evaluate it in real time."

[0633] In this way, the system of the present invention significantly improves the quality of vocational training by providing practice scenarios optimized for each user and generating immediate feedback. Furthermore, the introduction of an emotion engine provides detailed feedback according to the user's emotional state, enabling more practical and advanced skill improvement.

[0634] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0635] Step 1: Data collection

[0636] The server collects recorded audio data of sales proposals and presentations provided by companies. The server obtains audio data files from companies via HTTPS and stores them in data storage such as Google Cloud Storage. The input is the audio data file, and the output is the audio data file stored in the storage. The server converts the audio data into text data using the Google Cloud Speech-to-Text API and stores this text data in a database. The input is the audio data file, and the output is text data.

[0637] Step 2: Data Preprocessing

[0638] The server extracts features from text data and audio data. Features include voice stress, intonation patterns, speaking rate, etc. The Librosa library is used to extract features from the audio data and store them in a database. The input is audio data, and the output is feature data. The OpenCV library is used to analyze facial expression data, and facial expression features are also stored in the database. The input is facial expression image data, and the output is facial expression feature data.

[0639] Step 3: Model training

[0640] The server trains an AI model based on the collected feature data. Deep learning frameworks used include TensorFlow and PyTorch. For example, a speech analysis model is trained using TensorFlow and the model is saved. The input is the feature data, and the output is the trained AI model. Similarly, intonation analysis models and facial expression recognition models are trained and saved.

[0641] Step 4: Incorporating the Emotion Engine

[0642] The server incorporates an emotion engine into the system, adding the ability to recognize the user's emotional state. The emotion engine uses Affectiva or the Microsoft Emotion API. The API is used to analyze the user's emotions, and the results are stored in a database. The input is voice data and facial expression data, and the output is emotional state data.

[0643] Step 5: User authentication

[0644] A user logs in to the system from a terminal and enters authentication information. The terminal sends this information to the server, which then authenticates the user. OAuth2.0 or JWT is used for authentication. The input is the user authentication information, and the output is the authentication result.

[0645] Step 6: User Interaction

[0646] The device retrieves an individually tailored practice scenario from the server and presents it to the user. The user begins practicing based on the scenario displayed on the device. The device captures voice and facial expression data in real time and sends it to the server. The input is the practice scenario and the user's voice and facial expression data, and the output is the captured data.

[0647] Step 7: Generate feedback

[0648] The server analyzes the received voice and facial expression data in real time. The emotion engine also joins in to analyze the emotional state. For example, it analyzes the strength of the voice, changes in intonation, and recognizes facial expressions, and generates specific feedback. Based on the analysis results, it generates feedback such as "You should raise the tone of your voice in this part" or "If you smile briefly here, it will make a good impression." The input is the captured voice and facial expression data, and the output is the feedback data. The generated feedback is displayed to the user via the terminal.

[0649] The above is the specific processing flow of the program of this system.

[0650] (Application example 2)

[0651] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0652] Improving skills in sales, customer service, and presentations requires a lot of time and effort through manual education and training, but it is difficult to provide individually optimized feedback. There is also a lack of a system for evaluating the effectiveness of training in real time and immediately proposing improvement measures. Furthermore, it is difficult to maintain motivation for practice because feedback cannot take into account the emotional state of the trainee. This creates the problem of insufficient progress in improving customer service and presentation skills.

[0653] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice information, means for converting the voice information into character data, means for extracting features from the character data and the voice information, means for evaluating the user's skill level based on the extracted features, means for presenting an individually tailored practice scenario to the user, means for capturing the user's voice and facial expression information during practice and transmitting it to the server, means for analyzing the captured information in real time and generating feedback for the user, means for recognizing the user's emotional state, and means for presenting the generated feedback to the user. This makes it possible to effectively improve the user's skills through individually optimized practice scenarios and real-time feedback, and furthermore, by providing detailed feedback according to the user's emotional state, it is possible to achieve training effectiveness and maintain motivation.

[0654] "Voice information" refers to digitally recorded audio data, including the user's speech, tone of voice, intonation, and the like.

[0655] "Text data" refers to data that has been converted into text format by analyzing voice information, and is a written representation of what the user has said.

[0656] "Features" refer to measurable characteristics of a user's speech, such as stress, intonation, and speech rate, extracted from audio information and text data.

[0657] The "skill level" is an index that indicates the technical proficiency of the user's speaking style and facial expressions, evaluated based on the extracted feature amounts.

[0658] A "practice scenario" is an individually specialized practice scenario provided by the system for the purpose of improving the user's skills, and its content is customized according to the user's skill level.

[0659] "Facial expression information" refers to data that digitally records a user's facial expressions, including facial movements and emotional expressions.

[0660] The "emotional state" refers to the user's emotional state, which is recognized based on the user's facial expression information and voice information, and includes, for example, stress level and motivation level.

[0661] "Feedback" is specific advice about areas for improvement and effective practice methods for the user, generated based on analyzed voice information, text data, facial expression information, and emotional state.

[0662] "Real-time analysis" means processing the user's voice information and facial expression information instantly and obtaining analysis results instantly.

[0663] The present invention is an individually optimized dialogue simulator aimed at improving customer service skills in brick-and-mortar stores. This system involves communication between a server and a terminal, and provides users with optimized practice scenarios and feedback. A specific embodiment of this system is described below.

[0664] Data collection phase (server)

[0665] The server collects user voice information, which is uttered when the user practices customer service skills in a physical store and is recorded in real time using a smartphone or smart glasses.

[0666] Data preprocessing phase (server)

[0667] The collected speech information is converted into text data using automatic speech recognition (ASR) technology, such as Google Cloud Speech to Text. The Librosa library is then used to extract features from the speech data, including speech stress, intonation patterns, and speaking rate.

[0668] Model training phase (server)

[0669] The server trains AI models to assess the user's skill level based on the extracted features. These models include speech analysis, intonation analysis, and facial expression recognition. The trained models are saved and used to generate personalized practice scenarios for the user.

[0670] Emotion engine integration (server)

[0671] The server incorporates an emotion engine that recognizes the user's emotional state and analyzes it. The emotion engine is built using a deep learning library such as TensorFlow and evaluates the user's emotional state based on their voice and facial expression information.

[0672] User authentication phase (terminal)

[0673] The user logs in to the system from a terminal and begins practicing. The terminal can be a smartphone or smart glasses, and has the ability to capture the user's voice and facial expression information.

[0674] User Interaction Phase (Terminal)

[0675] The device retrieves an individually tailored practice scenario from the server and presents it to the user. The user begins practicing based on the displayed scenario, capturing voice and facial expression information in real time during the practice process and sending it to the server.

[0676] Feedback Phase (Server and Device)

[0677] The server analyzes the captured voice and facial expression information in real time and generates feedback. This feedback includes suggestions for improving voice stress, intonation, and facial expressions, as well as advice based on the user's emotional state. For example, the server provides specific feedback to the user, such as "It would be more effective if you raised the tone of your voice in this part," or "If you smile briefly here, you will make a good impression." The generated feedback is presented to the user via their device and can be used for future practice or when serving customers in physical stores.

[0678] Specific examples

[0679] For example, imagine a staff member introducing a new product. The staff member logs in to the system and starts a specialized practice scenario. The terminal captures the voice and facial expression information as the staff member explains the product's features and sends it to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the staff member's stress level and provides advice such as, "You may be too stressed and not able to concentrate on your practice. Take a break." The generated feedback is provided to the staff member via the terminal, and they use it to carry out their next practice.

[0680] Prompt Sentence Examples

[0681] Please generate a new customer service scenario. This scenario should be about a female customer in her twenties asking about a new product. The scenario should include the conversation with the customer, the information provided by the staff, and the features of the recommended product.

[0682] As described above, the system of the present invention supports the improvement of customer service skills in brick-and-mortar stores by providing practice scenarios optimized for each user and enabling real-time feedback. Furthermore, the introduction of an emotion engine allows for more detailed feedback based on the user's emotional state, enabling more effective training.

[0683] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0684] Step 1:

[0685] The server collects the user's voice information. This voice information is spoken by the user while practicing customer service skills and is recorded in real time using a device (smartphone or smart glasses). The input is the user's voice, and the output is digital voice data.

[0686] Step 2:

[0687] The server converts the collected voice information into text data using automatic speech recognition (ASR) technology. At this stage, the input is voice data and the output is text data. Specifically, an ASR service such as Google Cloud Speech to Text is used.

[0688] Step 3:

[0689] The server extracts features from text data and speech information. It uses the Librosa library to calculate features such as speech stress, intonation patterns, and speaking rate. The input is speech data and text data, and the output is a set of features.

[0690] Step 4:

[0691] The server evaluates the user's skill level based on the extracted features. This evaluation uses pre-trained AI models (voice analysis model, intonation analysis model, and facial expression recognition model). The input is the features, and the output is the skill level evaluation result.

[0692] Step 5:

[0693] The server generates a practice scenario specifically tailored for each user and sends it to the device. The scenario is customized based on the user's skill level assessment results. The input is the skill level assessment results, and the output is the practice scenario. The generative AI model generates the scenario according to the prompt sentence.

[0694] Step 6:

[0695] The terminal presents a practice scenario to the user, who then practices according to the scenario. The input is the practice scenario, and the output is the content of the practice performed by the user. Specifically, the scenario is displayed on the screen or guided by voice.

[0696] Step 7:

[0697] The device captures the user's voice and facial expression information while they practice and sends it to the server in real time. The input is the user's voice and facial expression information, and the output is the captured data. Specifically, the device's camera and microphone are always active.

[0698] Step 8:

[0699] The server analyzes the captured speech and facial expression information in real time and generates feedback for the user, including suggestions for improving intonation, vocal stress, facial expression, and emotional state. The input is the captured speech and facial expression information, and the output is the feedback.

[0700] Step 9:

[0701] The server uses an emotion engine to recognize the user's emotional state and reflects that information in the feedback. The input is captured voice and facial expression information, and the output is the evaluation result of the emotional state. Specifically, emotion analysis is performed using TensorFlow.

[0702] Step 10:

[0703] The terminal presents the generated feedback to the user. The input is the feedback, and the output is improvements or advice that is displayed or announced to the user by voice. Specific operations include displaying the feedback as text on the screen or announcing it by voice.

[0704] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0705] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0706] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0707] [Third embodiment]

[0708] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0709] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0710] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0711] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0712] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0713] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0714] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0715] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0716] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0717] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0718] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0719] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0720] The system and method of the present invention are an individually optimized dialogue simulator aimed at improving skills in sales, presentations, meetings, etc. This system collects and analyzes the user's voice data, provides individually tailored practice scenarios, and provides real-time feedback.

[0721] System program generation and processing details

[0722] 1. Data collection phase (server)

[0723] The server collects audio data and converts it into text. Audio recordings of specific sales proposals and presentations are collected from companies and stored as big data.

[0724] The collected voice data is converted into text data using automatic voice recognition technology.

[0725] 2. Data preprocessing phase (server)

[0726] The server extracts features from the text and audio data, such as speech stress, intonation patterns, and speaking rate.

[0727] 3. Model training phase (server)

[0728] Based on the extracted features, the server trains AI models to assess the user's skill level, including speech analysis models, intonation analysis models, and facial expression recognition models.

[0729] The trained model is saved and used to generate specialized training scenarios for the user.

[0730] 4. User authentication phase (terminal)

[0731] The user logs in to the system and enters authentication information, which the terminal sends to the server, where the user is identified through an authentication process.

[0732] 5. User Interaction Phase (Terminal)

[0733] The device retrieves a practice scenario specific to the user from the server and presents it to the user, who then begins practicing based on the scenario.

[0734] The device captures the user's voice and facial expression data while practicing and transmits it to the server in real time.

[0735] 6. Feedback Phase (Server and Device)

[0736] The server analyzes the received speech and facial expression data and generates real-time feedback to the user, including intonation corrections, improvements to specific expressions, and facial expression adjustments.

[0737] The device then presents the generated feedback to the user, providing specific advice such as, "It would be more effective if you raised the tone of your voice in this part," or "Smiling here will make a better impression."

[0738] Specific examples

[0739] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as "You should emphasize the intonation here" or "You might want to increase the tempo a little in this part." The generated feedback is provided to the user via the device, and the user uses it to carry out the next practice exercise.

[0740] In this way, the system of the present invention significantly improves the quality of vocational training and saves human resources by providing practice scenarios optimized for each user and enabling real-time feedback.In addition, by providing multifaceted feedback such as facial expressions and intonation, users can acquire practical and advanced skills.

[0741] The processing flow will be explained below.

[0742] Step 1:

[0743] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0744] Step 2:

[0745] The server converts the collected voice data into text using automatic speech recognition (ASR) technology, and the voice data is converted into text data and saved as text information.

[0746] Step 3:

[0747] The server extracts features from the text and audio data, calculating parameters such as voice stress, intonation patterns, and speaking rate.

[0748] Step 4:

[0749] The server trains AI models based on the preprocessed data to assess the user's skill level, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0750] Step 5:

[0751] Based on the results of the user's skill level assessment, the server generates specialized practice scenarios that provide specific content for the user to improve their skills.

[0752] Step 6:

[0753] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0754] Step 7:

[0755] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0756] Step 8:

[0757] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0758] Step 9:

[0759] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0760] Step 10:

[0761] The server generates specific feedback based on the analysis, such as "It would be more effective if you raised your voice tone in this part" or "If you smile briefly here, you'll make a better impression."

[0762] Step 11:

[0763] The device then presents the generated feedback to the user, who can use it to improve their next practice or presentation.

[0764] Step 12:

[0765] The user receives the provided feedback and practices the next time, adjusting the tone of their voice, intonation, and facial expressions based on the feedback.

[0766] The above is the specific flow of the program processing of the system aimed at improving user skills.

[0767] Example 1

[0768] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0769] Conventional vocational training systems have had the challenge of properly assessing individual users' skill levels and providing specialized practice scenarios and real-time feedback. Furthermore, it has been difficult to achieve both real-time performance and accuracy in the collection and analysis of voice and facial expression data. This has prevented users from improving their skills efficiently and effectively.

[0770] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0771] In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, means for extracting features from the text data and voice data, means for training a generative AI model that evaluates the user's skill level based on the extracted features, means for presenting an individually tailored practice scenario to the user, means for capturing the user's voice and facial expression data during practice and transmitting it to the server, means for analyzing the captured data in real time and generating feedback for the user, and means for presenting the generated feedback to the user. This makes it possible to accurately evaluate the user's skill level and provide individually optimized practice scenarios and real-time feedback.

[0772] "Voice data" refers to data that records in digital form the words spoken by a user.

[0773] "Text data" refers to data obtained by converting voice data into character information.

[0774] "Features" are important attributes or patterns that can be analyzed and extracted from speech or text data, such as speech stress, intonation patterns, and speaking rate.

[0775] "Skill level" is a standard for assessing a user's proficiency in a particular ability or technique.

[0776] A "generative AI model" is an artificial intelligence model that learns from training data based on features and evaluates and analyzes the user's skill level.

[0777] A "practice scenario" is a specific situation or task designed for users to practice, allowing them to practice in an environment that closely resembles a real-life work or presentation.

[0778] "Facial expression data" refers to data that captures a user's facial expressions and stores them in digital format.

[0779] "Feedback" refers to information that evaluates a user's performance during and after practice and provides suggestions and advice for improvement.

[0780] "Real-time analytics" refers to processing data and generating analytical results immediately as the data is collected.

[0781] "Capture" refers to recording the user's voice and facial expression data in digital format.

[0782] This invention is an individually optimized dialogue simulator aimed at improving skills in sales, presentations, meetings, etc. It collects and analyzes the user's voice data, provides individually specialized practice scenarios, and provides feedback in real time.

[0783] System configuration

[0784] The system is mainly composed of three components: a server, a terminal, and a user. The server collects data, analyzes it, trains models, and generates feedback. The terminal captures data and provides feedback. Users log in to the system, run practice scenarios, and receive feedback.

[0785] Hardware and software used

[0786] Server: Performs high-performance data analysis and model training using machine learning libraries such as TensorFlow and PyTorch, as well as the Google Cloud Speech-to-Text API.

[0787] Terminal: A device that captures voice and facial expressions and provides feedback to the user. The terminal software includes capture and real-time communication functions.

[0788] System Flow

[0789] 1. Collection and conversion of voice data (server)

[0790] The server collects audio data from sales proposals and presentations provided by companies and converts it into text data using the Google Cloud Speech-to-Text API. This process stores the audio as text information.

[0791] 2. Feature extraction (server)

[0792] The server extracts features such as voice intensity, intonation patterns, and speaking rate from the converted text and audio data using the Python librosa library.

[0793] 3. Training the AI ​​model (server)

[0794] Based on the extracted features, a generative AI model is trained using TensorFlow for speech analysis, intonation analysis, and facial expression recognition, and the model is saved after training.

[0795] 4. Presentation of practice scenario (terminal)

[0796] A user logs in to the system from a terminal and performs authentication. If authentication is successful, the server provides the user with a specialized practice scenario. The terminal displays this scenario, and the user begins practicing.

[0797] 5. Data capture and real-time transmission (terminal)

[0798] The device captures the user's voice and facial expression data while practicing and transmits it to the server in real time.

[0799] 6. Feedback Generation and Presentation (Server and Device)

[0800] The server analyzes the received data in real time and generates feedback, such as advice like "Raising your voice tone in this part will be more effective" or "Smiling here will make a better impression." The device then presents the generated feedback to the user.

[0801] Examples and prompts

[0802] For example, imagine a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as "You should emphasize the intonation here" or "It would be good to increase the tempo a little in this part." The generated feedback is provided to the user via the device, and the user uses this feedback to carry out the next practice session.

[0803] Example prompt sentence:

[0804] "Please give me feedback on where I should change my tone of voice when presenting a new product."

[0805] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0806] Step 1: Collecting voice data and converting it to text (server)

[0807] The server collects audio data of sales proposals and presentations provided by companies. The collected audio data is converted to text using the Google Cloud Speech-to-Text API. Specifically, the audio file is sent to the API, and the returned string is saved as a text file. The input is audio data, and the output is text data.

[0808] Step 2: Feature extraction (server)

[0809] The server extracts features from the converted text and audio data. Specifically, it uses the Python librosa library to analyze speech stress, intonation patterns, speaking rate, etc. The input is the text and audio data, and the output is the extracted feature data.

[0810] Step 3: Training the AI ​​model (server)

[0811] The server uses the extracted features to train a generative AI model. Specifically, it uses TensorFlow to create and train models for speech analysis, intonation analysis, and facial expression recognition. The input is the extracted feature data, and the output is the trained AI model.

[0812] Step 4: User authentication and access rights (device)

[0813] The user enters authentication information to log in to the system from their own terminal. The terminal sends this to the server, which executes the authentication process. The input is the user's authentication information, and the output is the authentication result (success or failure). If successful, access is granted, and if unsuccessful, an error message is returned.

[0814] Step 5: Acquire and present the practice scenario (terminal)

[0815] After successful authentication, the device retrieves a user-specific practice scenario from the server. This scenario is generated using a pre-trained AI model. The device displays the retrieved scenario to the user. The input is the successful user authentication status, and the output is the specialized practice scenario.

[0816] Step 6: Capture data during practice and send it in real time (on your device)

[0817] While the user practices based on the practice scenario, the device captures voice and facial expression data, which are then sent to the server in real time. The input is the user's voice and facial expression data, and the output is the captured data sent to the server.

[0818] Step 7: Generating and Presenting Feedback (Server and Device)

[0819] The server analyzes the received voice and facial expression data in real time and generates feedback. Specifically, it analyzes the user's intonation, speaking rate, facial expressions, etc. to identify areas for improvement. The generated feedback is sent to the device and presented to the user. The input is the captured data, and the output is the specific feedback content.

[0820] As a specific example of how it works, when a user gives a presentation, the server generates real-time feedback such as "It will be more effective if you raise the tone of your voice in this part" or "Smiling here will make a better impression," and provides it to the user via their device. Based on this feedback, the user can then practice the next part.

[0821] The above are the specific processing steps of the system.

[0822] (Application example 1)

[0823] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0824] There is a demand for a system that provides individually tailored practice scenarios based on the user's voice and facial expression data and provides feedback in real time. In customer service work, in particular, improving employee skills directly impacts sales and customer satisfaction, so effective and rapid training methods are essential. However, current systems face the challenge of making it difficult to provide detailed feedback in real time tailored to the user's individual skills and performance.

[0825] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0826] In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, and means for extracting features from the text data and voice data, thereby enabling the provision of practice scenarios based on individual user characteristics and real-time feedback.

[0827] "User voice data" refers to voice signals uttered by users, which are collected and analyzed to improve customer service skills.

[0828] "Means for converting to text" refers to technology that converts voice data into text information, and typically uses automatic speech recognition technology.

[0829] "Features" are patterns or attributes extracted from voice or text data and are used to evaluate a user's skills.

[0830] "Means for evaluating skill level" refers to technology that evaluates a user's ability and performance based on extracted features.

[0831] "Means for presenting practice scenarios" refers to technology that provides individually customized training content for the purpose of improving a user's skills.

[0832] "Voice and facial expression data" refers to data including the user's voice and facial expressions, and is used to analyze customer service performance.

[0833] "Means of real-time analysis" refers to technology that instantly processes and analyzes collected voice and facial expression data.

[0834] "Feedback" refers to improvements and advice provided to users based on the analysis results.

[0835] "Improving customer service skills" refers to techniques and methods aimed at improving a user's performance in customer service work.

[0836] This invention provides a training system for improving customer service skills. The system is configured by combining a smartphone terminal and a server, and collects user voice data and facial expression data, analyzing them in real time and providing feedback.

[0837] System Configuration

[0838] The system consists of the following main components:

[0839] 1. Data capture module (smartphone terminal)

[0840] Hardware: Smartphone camera, microphone.

[0841] Software: OpenCV (cv2), PyAudio. These are used to capture the user's voice and facial expression data in real time.

[0842] 2. Data transmission module (smartphone terminal)

[0843] Software: Requests library, used to send captured data to the server.

[0844] 3. Data analysis module (server)

[0845] Hardware: Cloud servers.

[0846] Software: TensorFlow, PyTorch, which are used to analyze the collected data and generate feedback.

[0847] 4. Feedback presentation module (smartphone device)

[0848] Software: A smartphone app that provides real-time feedback to users from the server.

[0849] Program processing

[0850] The smartphone device captures the user's voice and facial expressions and sends the data to a server. The captured data is acquired using the device's camera and microphone and processed by OpenCV and PyAudio. The data is then sent to the server using the Requests library.

[0851] The server analyzes the received voice and facial expression data using a generative AI model built with TensorFlow and PyTorch. This extracts features based on the user's intonation, vocal stress, speaking rate, and facial expressions, and evaluates the user's skill level.

[0852] Feedback is generated based on the analysis results from the server, providing specific advice such as "It will be more effective if you raise the tone of your voice in this part" or "Smiling here will make a better impression."

[0853] The smartphone device displays the feedback received from the server to the user in real time, allowing the user to immediately continue their training based on this feedback.

[0854] Examples and prompts

[0855] For example, when a sales clerk is explaining a new product to a customer, he or she places the smartphone in his or her breast pocket and launches the app. The app captures the clerk's voice and facial expressions and sends them to a server. The server analyzes the clerk's speech rate and facial expression patterns and provides real-time feedback such as, "You should smile more and speak more slowly when explaining the product's features."

[0856] Prompt Sentence Examples

[0857] "Please explain the features of our new product. Could you please give us some advice on what would be an effective explanation?"

[0858] "Please check the flow of conversations in real time while serving customers and let us know where we can improve."

[0859] "After you've finished explaining the product, please give us your feedback on what needs to be improved."

[0860] This is expected to enable efficient improvement of customer service skills and ultimately contribute to increased customer satisfaction.

[0861] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0862] Step 1: Data Capture

[0863] The smartphone device allows the user to perform a specific scenario (e.g., a new product presentation), and uses the device's camera and microphone to capture the user's voice and facial expression data in real time. This data is then sent raw to the next processing step.

[0864] Input: User's voice data, user's facial expression data

[0865] Output: Raw audio and facial expression data

[0866] Step 2: Send data

[0867] The captured voice and facial expression data is sent to the server using the smartphone's Requests library, where it is encoded and sent as an HTTP request.

[0868] Input: Raw audio and facial expression data

[0869] Output: The encoded data sent to the server.

[0870] Step 3: Convert audio data to text

[0871] The server converts the received voice data into text using automatic speech recognition (ASR), which converts the voice data into text data, making the voice content available as text.

[0872] Input: Encoded audio data

[0873] Output: Text data

[0874] Step 4: Feature extraction

[0875] The server extracts features from the text and audio data, such as speech stress, intonation, speaking rate, and text content, using speech analysis models and natural language processing (NLP) models.

[0876] Input: Text data, audio data

[0877] Output: Extracted features

[0878] Step 5: Assess your skill level

[0879] The server evaluates the user's skill level based on the extracted features, using a generative AI model to represent the user's customer service skills and presentation quality numerically and graphically.

[0880] Input: Extracted features

[0881] Output: User's skill level

[0882] Step 6: Provide a practice scenario

[0883] The server generates personalized practice scenarios based on the user's skill level, taking into account the user's weaknesses and areas for improvement, and suggests new scenarios.

[0884] Input: User's skill level

[0885] Output: A customized training scenario

[0886] Step 7: Capture user practice data

[0887] The user practices according to the provided practice scenario, while the smartphone again captures voice and facial expression data and sends them to the server. At the same time, the current practice status is monitored in real time.

[0888] Input: Voice data during practice, facial expression data during practice

[0889] Output: Practice data sent in real time

[0890] Step 8: Generate real-time feedback

[0891] The server analyzes the received practice data in real time and generates feedback for the user, including corrections to intonation, improvements to specific expressions, and adjustments to facial expressions.

[0892] Input: Voice data during practice, facial expression data during practice

[0893] Output: Real-time feedback

[0894] Step 9: Provide feedback

[0895] The smartphone device receives feedback from the server and presents it to the user in text, audio, visual, or other formats, allowing the user to use the feedback in their next practice.

[0896] Input: Real-time feedback

[0897] Output: Feedback presented to the user

[0898] This series of steps makes it possible to efficiently support users in improving their customer service skills.

[0899] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0900] This invention is a personalized dialogue simulator designed to improve skills in sales, presentations, meetings, and other areas. The system collects and analyzes the user's voice and facial expression data, provides customized practice scenarios, and provides real-time feedback. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it provides deeper feedback.

[0901] System program generation and processing details

[0902] 1. Data collection phase (server)

[0903] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0904] The collected voice data is converted into text using automatic speech recognition (ASR) technology and saved as text information.

[0905] 2. Data preprocessing phase (server)

[0906] The server extracts features from the text and audio data, including parameters such as voice stress, intonation patterns, and speaking rate.

[0907] 3. Model training phase (server)

[0908] The server trains AI models to assess the user's skill level based on the extracted features, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0909] The trained model is saved and used to generate specialized training scenarios for the user.

[0910] 4. Emotion engine integration (server)

[0911] The server incorporates a new emotion engine, which adds the ability to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation.

[0912] 5. User authentication phase (terminal)

[0913] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0914] 6. User Interaction Phase (Terminal)

[0915] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0916] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0917] 7. Feedback Phase (Server and Device)

[0918] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0919] The emotion engine recognizes the user's emotional state and assesses their stress level and motivation, which are then used to generate further feedback.

[0920] The server generates specific feedback based on the analysis, such as "You'll be more effective if you raise your voice tone here," "A quick smile here will make a better impression," or "You're under a lot of stress, so consider slowing down your practice."

[0921] The device displays the generated feedback to the user and provides specific instructions on how to improve for the next practice or presentation.

[0922] Specific examples

[0923] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the user's stress level and provides advice such as, "You may be too stressed and unable to concentrate on your practice. Take a break." The generated feedback is provided to the user via the device, and the user uses this information to carry out the next practice.

[0924] In this way, the system of the present invention significantly improves the quality of vocational training by providing practice scenarios optimized for each user and enabling real-time feedback. Furthermore, the introduction of an emotion engine provides detailed feedback according to the user's emotional state, enabling more practical and advanced skill improvement.

[0925] The processing flow will be explained below.

[0926] Step 1:

[0927] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[0928] Step 2:

[0929] The server converts the collected voice data into text using automatic speech recognition (ASR) technology, and the voice data is converted into text data and saved as text information.

[0930] Step 3:

[0931] The server extracts features from the text and audio data, calculating parameters such as voice stress, intonation patterns, and speaking rate.

[0932] Step 4:

[0933] The server trains AI models based on the preprocessed data to assess the user's skill level, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[0934] Step 5:

[0935] The server incorporates a new emotion engine, which adds the ability to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation.

[0936] Step 6:

[0937] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[0938] Step 7:

[0939] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[0940] Step 8:

[0941] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[0942] Step 9:

[0943] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[0944] Step 10:

[0945] The server generates specific feedback based on the analysis, such as "It would be more effective if you raised your voice tone in this part" or "If you smile briefly here, you'll make a better impression."

[0946] Step 11:

[0947] The emotion engine recognizes the user's emotional state from their voice and facial expression data and evaluates their stress level and motivation. For example, if the user's voice tone is low and their facial expression is stiff, it may determine that the user is feeling stressed.

[0948] Step 12:

[0949] The server generates further feedback based on the emotion engine's evaluation, such as advice like, "You may be under a lot of stress and not be able to concentrate on your practice. Take a break."

[0950] Step 13:

[0951] The device displays the generated feedback to the user and provides specific instructions on how to improve in the next practice or presentation. Based on the feedback, the user can adjust their voice tone, intonation, and facial expressions.

[0952] Step 14:

[0953] The user receives the provided feedback and practices the next time. Based on the feedback, the user can adjust the tone, intonation, and facial expressions of their voice to maximize the effectiveness of their training.

[0954] Example 2

[0955] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0956] Conventional vocational training systems have difficulty providing feedback that fully considers the characteristics and emotional state of individual users. This makes it difficult to efficiently improve users' practical skills. In addition, real-time feedback generation is often delayed, making it difficult to provide immediate improvements.

[0957] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, means for extracting features from the text data and voice data, means for evaluating the user's skill level based on the extracted features, means for recognizing the user's emotional state, means for presenting a practice scenario individually tailored to the user, means for capturing voice and facial expression data of the user during practice and transmitting the captured data to the server, means for analyzing the captured data in real time and generating feedback for the user, and means for presenting the generated feedback to the user. This makes it possible to provide a practice scenario individually optimized for the user and generate feedback instantly.

[0958] "Voice data" is digital data that records what the user has said.

[0959] "Text data" refers to character information generated from voice data using automatic voice recognition technology.

[0960] "Features" are parameters extracted from speech and facial expression data and used for analysis and model training. Specifically, they include speech stress, intonation patterns, speaking rate, etc.

[0961] The "means for assessing skill level" refers to an algorithm or model for assessing a user's ability or performance based on the extracted features.

[0962] The "means for recognizing emotional state" refers to an engine or technology that analyzes the user's voice data and facial expression data to grasp the user's emotional state, such as stress and motivation level.

[0963] The "means for presenting a practice scenario" is an interface or system for presenting practice content that is specialized for each individual user.

[0964] The "capturing means" refers to a device or software for collecting voice and facial expression data in real time while the user is practicing.

[0965] The "means for generating feedback" refers to an algorithm or system that analyzes the captured voice and facial expression data and presents areas for improvement to the user.

[0966] A "means for presenting feedback" is an interface or device that visually or audibly notifies the user of the generated feedback.

[0967] This invention is a personalized dialogue simulator designed to improve skills in sales, presentations, meetings, and other areas. The system collects and analyzes the user's voice and facial expression data, provides customized practice scenarios, and provides real-time feedback. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it provides deeper feedback.

[0968] First, the server collects recorded audio data of sales proposals and presentations provided by companies. This audio data is stored on the server as big data. The collected audio data is converted into text using automatic speech recognition (ASR) technology and saved as text information. A variety of automatic speech recognition APIs can be used for the ASR technology used.

[0969] The server then extracts features from the text and audio data, such as voice stress, intonation patterns, and speaking rate. A speech signal processing library is used to extract the features, and a computer vision library is used to analyze facial expression data.

[0970] The server then trains AI models based on these feature data. The trained models include a speech analysis model, an intonation analysis model, and a facial expression recognition model. The deep learning frameworks used include commonly used deep learning platforms.

[0971] The server also incorporates an emotion engine to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation. An emotion recognition API is available for this emotion analysis.

[0972] The user logs in to the system from their device and enters their authentication information. The device then sends this information to the server for user authentication. If authentication is successful, the device retrieves an individually tailored practice scenario from the server and presents it to the user. The user then begins practicing based on the displayed scenario. The device captures the user's voice and facial expression data in real time while they practice and sends the data to the server.

[0973] The server analyzes the received voice and facial expression data in real time. Analysis includes voice stress, changes in intonation, and facial expression recognition. The emotion engine recognizes the user's emotional state and evaluates their stress level and motivation. This evaluation result is used to generate further feedback. Based on the analysis results, the server generates specific feedback, such as "You should raise the tone of your voice in this part," "A quick smile here will create a good impression," or "You are under high stress, so consider slowing down your practice pace." The generated feedback is displayed to the user via their device, providing specific instructions on how to improve in their next practice or presentation.

[0974] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the user's stress level and provides advice such as, "You may be too stressed and unable to concentrate on your practice. Take a break." The generated feedback is provided to the user via the device, and the user uses this information to carry out the next practice.

[0975] Examples of input prompts include, "I would like the system to give me specific advice on what to pay attention to when presenting a new product." and "Please capture my voice and facial expression data during the presentation and evaluate it in real time."

[0976] In this way, the system of the present invention significantly improves the quality of vocational training by providing practice scenarios optimized for each user and generating immediate feedback. Furthermore, the introduction of an emotion engine provides detailed feedback according to the user's emotional state, enabling more practical and advanced skill improvement.

[0977] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0978] Step 1: Data collection

[0979] The server collects recorded audio data of sales proposals and presentations provided by companies. The server obtains audio data files from companies via HTTPS and stores them in data storage such as Google Cloud Storage. The input is the audio data file, and the output is the audio data file stored in the storage. The server converts the audio data into text data using the Google Cloud Speech-to-Text API and stores this text data in a database. The input is the audio data file, and the output is text data.

[0980] Step 2: Data Preprocessing

[0981] The server extracts features from text data and audio data. Features include voice stress, intonation patterns, speaking rate, etc. The Librosa library is used to extract features from the audio data and store them in a database. The input is audio data, and the output is feature data. The OpenCV library is used to analyze facial expression data, and facial expression features are also stored in the database. The input is facial expression image data, and the output is facial expression feature data.

[0982] Step 3: Model training

[0983] The server trains an AI model based on the collected feature data. Deep learning frameworks used include TensorFlow and PyTorch. For example, a speech analysis model is trained using TensorFlow and the model is saved. The input is the feature data, and the output is the trained AI model. Similarly, intonation analysis models and facial expression recognition models are trained and saved.

[0984] Step 4: Incorporating the Emotion Engine

[0985] The server incorporates an emotion engine into the system, adding the ability to recognize the user's emotional state. The emotion engine uses Affectiva or the Microsoft Emotion API. The API is used to analyze the user's emotions, and the results are stored in a database. The input is voice data and facial expression data, and the output is emotional state data.

[0986] Step 5: User authentication

[0987] A user logs in to the system from a terminal and enters authentication information. The terminal sends this information to the server, which then authenticates the user. OAuth2.0 or JWT is used for authentication. The input is the user authentication information, and the output is the authentication result.

[0988] Step 6: User Interaction

[0989] The device retrieves an individually tailored practice scenario from the server and presents it to the user. The user begins practicing based on the scenario displayed on the device. The device captures voice and facial expression data in real time and sends it to the server. The input is the practice scenario and the user's voice and facial expression data, and the output is the captured data.

[0990] Step 7: Generate feedback

[0991] The server analyzes the received voice and facial expression data in real time. The emotion engine also joins in to analyze the emotional state. For example, it analyzes the strength of the voice, changes in intonation, and recognizes facial expressions, and generates specific feedback. Based on the analysis results, it generates feedback such as "You should raise the tone of your voice in this part" or "If you smile briefly here, it will make a good impression." The input is the captured voice and facial expression data, and the output is the feedback data. The generated feedback is displayed to the user via the terminal.

[0992] The above is the specific processing flow of the program of this system.

[0993] (Application example 2)

[0994] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0995] Improving skills in sales, customer service, and presentations requires a lot of time and effort through manual education and training, but it is difficult to provide individually optimized feedback. There is also a lack of a system for evaluating the effectiveness of training in real time and immediately proposing improvement measures. Furthermore, it is difficult to maintain motivation for practice because feedback cannot take into account the emotional state of the trainee. This creates the problem of insufficient progress in improving customer service and presentation skills.

[0996] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice information, means for converting the voice information into character data, means for extracting features from the character data and the voice information, means for evaluating the user's skill level based on the extracted features, means for presenting an individually tailored practice scenario to the user, means for capturing the user's voice and facial expression information during practice and transmitting it to the server, means for analyzing the captured information in real time and generating feedback for the user, means for recognizing the user's emotional state, and means for presenting the generated feedback to the user. This makes it possible to effectively improve the user's skills through individually optimized practice scenarios and real-time feedback, and furthermore, by providing detailed feedback according to the user's emotional state, it is possible to achieve training effectiveness and maintain motivation.

[0997] "Voice information" refers to digitally recorded audio data, including the user's speech, tone of voice, intonation, and the like.

[0998] "Text data" refers to data that has been converted into text format by analyzing voice information, and is a written representation of what the user has said.

[0999] "Features" refer to measurable characteristics of a user's speech, such as stress, intonation, and speech rate, extracted from audio information and text data.

[1000] The "skill level" is an index that indicates the technical proficiency of the user's speaking style and facial expressions, evaluated based on the extracted feature amounts.

[1001] A "practice scenario" is an individually specialized practice scenario provided by the system for the purpose of improving the user's skills, and its content is customized according to the user's skill level.

[1002] "Facial expression information" refers to data that digitally records a user's facial expressions, including facial movements and emotional expressions.

[1003] The "emotional state" refers to the user's emotional state, which is recognized based on the user's facial expression information and voice information, and includes, for example, stress level and motivation level.

[1004] "Feedback" is specific advice about areas for improvement and effective practice methods for the user, generated based on analyzed voice information, text data, facial expression information, and emotional state.

[1005] "Real-time analysis" means processing the user's voice information and facial expression information instantly and obtaining analysis results instantly.

[1006] The present invention is an individually optimized dialogue simulator aimed at improving customer service skills in brick-and-mortar stores. This system involves communication between a server and a terminal, and provides users with optimized practice scenarios and feedback. A specific embodiment of this system is described below.

[1007] Data collection phase (server)

[1008] The server collects user voice information, which is uttered when the user practices customer service skills in a physical store and is recorded in real time using a smartphone or smart glasses.

[1009] Data preprocessing phase (server)

[1010] The collected speech information is converted into text data using automatic speech recognition (ASR) technology, such as Google Cloud Speech to Text. The Librosa library is then used to extract features from the speech data, including speech stress, intonation patterns, and speaking rate.

[1011] Model training phase (server)

[1012] The server trains AI models to assess the user's skill level based on the extracted features. These models include speech analysis, intonation analysis, and facial expression recognition. The trained models are saved and used to generate personalized practice scenarios for the user.

[1013] Emotion engine integration (server)

[1014] The server incorporates an emotion engine that recognizes the user's emotional state and analyzes it. The emotion engine is built using a deep learning library such as TensorFlow and evaluates the user's emotional state based on their voice and facial expression information.

[1015] User authentication phase (terminal)

[1016] The user logs in to the system from a terminal and begins practicing. The terminal can be a smartphone or smart glasses, and has the ability to capture the user's voice and facial expression information.

[1017] User Interaction Phase (Terminal)

[1018] The device retrieves an individually tailored practice scenario from the server and presents it to the user. The user begins practicing based on the displayed scenario, capturing voice and facial expression information in real time during the practice process and sending it to the server.

[1019] Feedback Phase (Server and Device)

[1020] The server analyzes the captured voice and facial expression information in real time and generates feedback. This feedback includes suggestions for improving voice stress, intonation, and facial expressions, as well as advice based on the user's emotional state. For example, the server provides specific feedback to the user, such as "It would be more effective if you raised the tone of your voice in this part," or "If you smile briefly here, you will make a good impression." The generated feedback is presented to the user via their device and can be used for future practice or when serving customers in physical stores.

[1021] Specific examples

[1022] For example, imagine a staff member introducing a new product. The staff member logs in to the system and starts a specialized practice scenario. The terminal captures the voice and facial expression information as the staff member explains the product's features and sends it to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the staff member's stress level and provides advice such as, "You may be too stressed and not able to concentrate on your practice. Take a break." The generated feedback is provided to the staff member via the terminal, and they use it to carry out their next practice.

[1023] Prompt Sentence Examples

[1024] Please generate a new customer service scenario. This scenario should be about a female customer in her twenties asking about a new product. The scenario should include the conversation with the customer, the information provided by the staff, and the features of the recommended product.

[1025] As described above, the system of the present invention supports the improvement of customer service skills in brick-and-mortar stores by providing practice scenarios optimized for each user and enabling real-time feedback. Furthermore, the introduction of an emotion engine allows for more detailed feedback based on the user's emotional state, enabling more effective training.

[1026] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1027] Step 1:

[1028] The server collects the user's voice information. This voice information is spoken by the user while practicing customer service skills and is recorded in real time using a device (smartphone or smart glasses). The input is the user's voice, and the output is digital voice data.

[1029] Step 2:

[1030] The server converts the collected voice information into text data using automatic speech recognition (ASR) technology. At this stage, the input is voice data and the output is text data. Specifically, an ASR service such as Google Cloud Speech to Text is used.

[1031] Step 3:

[1032] The server extracts features from text data and speech information. It uses the Librosa library to calculate features such as speech stress, intonation patterns, and speaking rate. The input is speech data and text data, and the output is a set of features.

[1033] Step 4:

[1034] The server evaluates the user's skill level based on the extracted features. This evaluation uses pre-trained AI models (voice analysis model, intonation analysis model, and facial expression recognition model). The input is the features, and the output is the skill level evaluation result.

[1035] Step 5:

[1036] The server generates a practice scenario specifically tailored for each user and sends it to the device. The scenario is customized based on the user's skill level assessment results. The input is the skill level assessment results, and the output is the practice scenario. The generative AI model generates the scenario according to the prompt sentence.

[1037] Step 6:

[1038] The terminal presents a practice scenario to the user, who then practices according to the scenario. The input is the practice scenario, and the output is the content of the practice performed by the user. Specifically, the scenario is displayed on the screen or guided by voice.

[1039] Step 7:

[1040] The device captures the user's voice and facial expression information while they practice and sends it to the server in real time. The input is the user's voice and facial expression information, and the output is the captured data. Specifically, the device's camera and microphone are always active.

[1041] Step 8:

[1042] The server analyzes the captured speech and facial expression information in real time and generates feedback for the user, including suggestions for improving intonation, vocal stress, facial expression, and emotional state. The input is the captured speech and facial expression information, and the output is the feedback.

[1043] Step 9:

[1044] The server uses an emotion engine to recognize the user's emotional state and reflects that information in the feedback. The input is captured voice and facial expression information, and the output is the evaluation result of the emotional state. Specifically, emotion analysis is performed using TensorFlow.

[1045] Step 10:

[1046] The terminal presents the generated feedback to the user. The input is the feedback, and the output is improvements or advice that is displayed or announced to the user by voice. Specific operations include displaying the feedback as text on the screen or announcing it by voice.

[1047] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1048] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1049] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1050] [Fourth embodiment]

[1051] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1052] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1053] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1054] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1055] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1056] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1057] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1058] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1059] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1060] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1061] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1062] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1063] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1064] The system and method of the present invention are an individually optimized dialogue simulator aimed at improving skills in sales, presentations, meetings, etc. This system collects and analyzes the user's voice data, provides individually tailored practice scenarios, and provides real-time feedback.

[1065] System program generation and processing details

[1066] 1. Data collection phase (server)

[1067] The server collects audio data and converts it into text. Audio recordings of specific sales proposals and presentations are collected from companies and stored as big data.

[1068] The collected voice data is converted into text data using automatic voice recognition technology.

[1069] 2. Data preprocessing phase (server)

[1070] The server extracts features from the text and audio data, such as speech stress, intonation patterns, and speaking rate.

[1071] 3. Model training phase (server)

[1072] Based on the extracted features, the server trains AI models to assess the user's skill level, including speech analysis models, intonation analysis models, and facial expression recognition models.

[1073] The trained model is saved and used to generate specialized training scenarios for the user.

[1074] 4. User authentication phase (terminal)

[1075] The user logs in to the system and enters authentication information, which the terminal sends to the server, where the user is identified through an authentication process.

[1076] 5. User Interaction Phase (Terminal)

[1077] The device retrieves a practice scenario specific to the user from the server and presents it to the user, who then begins practicing based on the scenario.

[1078] The device captures the user's voice and facial expression data while practicing and transmits it to the server in real time.

[1079] 6. Feedback Phase (Server and Device)

[1080] The server analyzes the received speech and facial expression data and generates real-time feedback to the user, including intonation corrections, improvements to specific expressions, and facial expression adjustments.

[1081] The device then presents the generated feedback to the user, providing specific advice such as, "It would be more effective if you raised the tone of your voice in this part," or "Smiling here will make a better impression."

[1082] Specific examples

[1083] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as "You should emphasize the intonation here" or "You might want to increase the tempo a little in this part." The generated feedback is provided to the user via the device, and the user uses it to carry out the next practice exercise.

[1084] In this way, the system of the present invention significantly improves the quality of vocational training and saves human resources by providing practice scenarios optimized for each user and enabling real-time feedback.In addition, by providing multifaceted feedback such as facial expressions and intonation, users can acquire practical and advanced skills.

[1085] The processing flow will be explained below.

[1086] Step 1:

[1087] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[1088] Step 2:

[1089] The server converts the collected voice data into text using automatic speech recognition (ASR) technology, and the voice data is converted into text data and saved as text information.

[1090] Step 3:

[1091] The server extracts features from the text and audio data, calculating parameters such as voice stress, intonation patterns, and speaking rate.

[1092] Step 4:

[1093] The server trains AI models based on the preprocessed data to assess the user's skill level, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[1094] Step 5:

[1095] Based on the results of the user's skill level assessment, the server generates specialized practice scenarios that provide specific content for the user to improve their skills.

[1096] Step 6:

[1097] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[1098] Step 7:

[1099] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[1100] Step 8:

[1101] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[1102] Step 9:

[1103] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[1104] Step 10:

[1105] The server generates specific feedback based on the analysis, such as "It would be more effective if you raised your voice tone in this part" or "If you smile briefly here, you'll make a better impression."

[1106] Step 11:

[1107] The device then presents the generated feedback to the user, who can use it to improve their next practice or presentation.

[1108] Step 12:

[1109] The user receives the provided feedback and practices the next time, adjusting the tone of their voice, intonation, and facial expressions based on the feedback.

[1110] The above is the specific flow of the program processing of the system aimed at improving user skills.

[1111] Example 1

[1112] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1113] Conventional vocational training systems have had the challenge of properly assessing individual users' skill levels and providing specialized practice scenarios and real-time feedback. Furthermore, it has been difficult to achieve both real-time performance and accuracy in the collection and analysis of voice and facial expression data. This has prevented users from improving their skills efficiently and effectively.

[1114] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1115] In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, means for extracting features from the text data and voice data, means for training a generative AI model that evaluates the user's skill level based on the extracted features, means for presenting an individually tailored practice scenario to the user, means for capturing the user's voice and facial expression data during practice and transmitting it to the server, means for analyzing the captured data in real time and generating feedback for the user, and means for presenting the generated feedback to the user. This makes it possible to accurately evaluate the user's skill level and provide individually optimized practice scenarios and real-time feedback.

[1116] "Voice data" refers to data that records in digital form the words spoken by a user.

[1117] "Text data" refers to data obtained by converting voice data into character information.

[1118] "Features" are important attributes or patterns that can be analyzed and extracted from speech or text data, such as speech stress, intonation patterns, and speaking rate.

[1119] "Skill level" is a standard for assessing a user's proficiency in a particular ability or technique.

[1120] A "generative AI model" is an artificial intelligence model that learns from training data based on features and evaluates and analyzes the user's skill level.

[1121] A "practice scenario" is a specific situation or task designed for users to practice, allowing them to practice in an environment that closely resembles a real-life work or presentation.

[1122] "Facial expression data" refers to data that captures a user's facial expressions and stores them in digital format.

[1123] "Feedback" refers to information that evaluates a user's performance during and after practice and provides suggestions and advice for improvement.

[1124] "Real-time analytics" refers to processing data and generating analytical results immediately as the data is collected.

[1125] "Capture" refers to recording the user's voice and facial expression data in digital format.

[1126] This invention is an individually optimized dialogue simulator aimed at improving skills in sales, presentations, meetings, etc. It collects and analyzes the user's voice data, provides individually specialized practice scenarios, and provides feedback in real time.

[1127] System configuration

[1128] The system is mainly composed of three components: a server, a terminal, and a user. The server collects data, analyzes it, trains models, and generates feedback. The terminal captures data and provides feedback. Users log in to the system, run practice scenarios, and receive feedback.

[1129] Hardware and software used

[1130] Server: Performs high-performance data analysis and model training using machine learning libraries such as TensorFlow and PyTorch, as well as the Google Cloud Speech-to-Text API.

[1131] Terminal: A device that captures voice and facial expressions and provides feedback to the user. The terminal software includes capture and real-time communication functions.

[1132] System Flow

[1133] 1. Collection and conversion of voice data (server)

[1134] The server collects audio data from sales proposals and presentations provided by companies and converts it into text data using the Google Cloud Speech-to-Text API. This process stores the audio as text information.

[1135] 2. Feature extraction (server)

[1136] The server extracts features such as voice intensity, intonation patterns, and speaking rate from the converted text and audio data using the Python librosa library.

[1137] 3. Training the AI ​​model (server)

[1138] Based on the extracted features, a generative AI model is trained using TensorFlow for speech analysis, intonation analysis, and facial expression recognition, and the model is saved after training.

[1139] 4. Presentation of practice scenario (terminal)

[1140] A user logs in to the system from a terminal and performs authentication. If authentication is successful, the server provides the user with a specialized practice scenario. The terminal displays this scenario, and the user begins practicing.

[1141] 5. Data capture and real-time transmission (terminal)

[1142] The device captures the user's voice and facial expression data while practicing and transmits it to the server in real time.

[1143] 6. Feedback Generation and Presentation (Server and Device)

[1144] The server analyzes the received data in real time and generates feedback, such as advice like "Raising your voice tone in this part will be more effective" or "Smiling here will make a better impression." The device then presents the generated feedback to the user.

[1145] Examples and prompts

[1146] For example, imagine a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as "You should emphasize the intonation here" or "It would be good to increase the tempo a little in this part." The generated feedback is provided to the user via the device, and the user uses this feedback to carry out the next practice session.

[1147] Example prompt sentence:

[1148] "Please give me feedback on where I should change my tone of voice when presenting a new product."

[1149] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1150] Step 1: Collecting voice data and converting it to text (server)

[1151] The server collects audio data of sales proposals and presentations provided by companies. The collected audio data is converted to text using the Google Cloud Speech-to-Text API. Specifically, the audio file is sent to the API, and the returned string is saved as a text file. The input is audio data, and the output is text data.

[1152] Step 2: Feature extraction (server)

[1153] The server extracts features from the converted text and audio data. Specifically, it uses the Python librosa library to analyze speech stress, intonation patterns, speaking rate, etc. The input is the text and audio data, and the output is the extracted feature data.

[1154] Step 3: Training the AI ​​model (server)

[1155] The server uses the extracted features to train a generative AI model. Specifically, it uses TensorFlow to create and train models for speech analysis, intonation analysis, and facial expression recognition. The input is the extracted feature data, and the output is the trained AI model.

[1156] Step 4: User authentication and access rights (device)

[1157] The user enters authentication information to log in to the system from their own terminal. The terminal sends this to the server, which executes the authentication process. The input is the user's authentication information, and the output is the authentication result (success or failure). If successful, access is granted, and if unsuccessful, an error message is returned.

[1158] Step 5: Acquire and present the practice scenario (terminal)

[1159] After successful authentication, the device retrieves a user-specific practice scenario from the server. This scenario is generated using a pre-trained AI model. The device displays the retrieved scenario to the user. The input is the successful user authentication status, and the output is the specialized practice scenario.

[1160] Step 6: Capture data during practice and send it in real time (on your device)

[1161] While the user practices based on the practice scenario, the device captures voice and facial expression data, which are then sent to the server in real time. The input is the user's voice and facial expression data, and the output is the captured data sent to the server.

[1162] Step 7: Generating and Presenting Feedback (Server and Device)

[1163] The server analyzes the received voice and facial expression data in real time and generates feedback. Specifically, it analyzes the user's intonation, speaking rate, facial expressions, etc. to identify areas for improvement. The generated feedback is sent to the device and presented to the user. The input is the captured data, and the output is the specific feedback content.

[1164] As a specific example of how it works, when a user gives a presentation, the server generates real-time feedback such as "It will be more effective if you raise the tone of your voice in this part" or "Smiling here will make a better impression," and provides it to the user via their device. Based on this feedback, the user can then practice the next part.

[1165] The above are the specific processing steps of the system.

[1166] (Application example 1)

[1167] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1168] There is a demand for a system that provides individually tailored practice scenarios based on the user's voice and facial expression data and provides feedback in real time. In customer service work, in particular, improving employee skills directly impacts sales and customer satisfaction, so effective and rapid training methods are essential. However, current systems face the challenge of making it difficult to provide detailed feedback in real time tailored to the user's individual skills and performance.

[1169] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1170] In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, and means for extracting features from the text data and voice data, thereby enabling the provision of practice scenarios based on individual user characteristics and real-time feedback.

[1171] "User voice data" refers to voice signals uttered by users, which are collected and analyzed to improve customer service skills.

[1172] "Means for converting to text" refers to technology that converts voice data into text information, and typically uses automatic speech recognition technology.

[1173] "Features" are patterns or attributes extracted from voice or text data and are used to evaluate a user's skills.

[1174] "Means for evaluating skill level" refers to technology that evaluates a user's ability and performance based on extracted features.

[1175] "Means for presenting practice scenarios" refers to technology that provides individually customized training content for the purpose of improving a user's skills.

[1176] "Voice and facial expression data" refers to data including the user's voice and facial expressions, and is used to analyze customer service performance.

[1177] "Means of real-time analysis" refers to technology that instantly processes and analyzes collected voice and facial expression data.

[1178] "Feedback" refers to improvements and advice provided to users based on the analysis results.

[1179] "Improving customer service skills" refers to techniques and methods aimed at improving a user's performance in customer service work.

[1180] This invention provides a training system for improving customer service skills. The system is configured by combining a smartphone terminal and a server, and collects user voice data and facial expression data, analyzing them in real time and providing feedback.

[1181] System Configuration

[1182] The system consists of the following main components:

[1183] 1. Data capture module (smartphone terminal)

[1184] Hardware: Smartphone camera, microphone.

[1185] Software: OpenCV (cv2), PyAudio. These are used to capture the user's voice and facial expression data in real time.

[1186] 2. Data transmission module (smartphone terminal)

[1187] Software: Requests library, used to send captured data to the server.

[1188] 3. Data analysis module (server)

[1189] Hardware: Cloud servers.

[1190] Software: TensorFlow, PyTorch, which are used to analyze the collected data and generate feedback.

[1191] 4. Feedback presentation module (smartphone device)

[1192] Software: A smartphone app that provides real-time feedback to users from the server.

[1193] Program processing

[1194] The smartphone device captures the user's voice and facial expressions and sends the data to a server. The captured data is acquired using the device's camera and microphone and processed by OpenCV and PyAudio. The data is then sent to the server using the Requests library.

[1195] The server analyzes the received voice and facial expression data using a generative AI model built with TensorFlow and PyTorch. This extracts features based on the user's intonation, vocal stress, speaking rate, and facial expressions, and evaluates the user's skill level.

[1196] Feedback is generated based on the analysis results from the server, providing specific advice such as "It will be more effective if you raise the tone of your voice in this part" or "Smiling here will make a better impression."

[1197] The smartphone device displays the feedback received from the server to the user in real time, allowing the user to immediately continue their training based on this feedback.

[1198] Examples and prompts

[1199] For example, when a sales clerk is explaining a new product to a customer, he or she places the smartphone in his or her breast pocket and launches the app. The app captures the clerk's voice and facial expressions and sends them to a server. The server analyzes the clerk's speech rate and facial expression patterns and provides real-time feedback such as, "You should smile more and speak more slowly when explaining the product's features."

[1200] Prompt Sentence Examples

[1201] "Please explain the features of our new product. Could you please give us some advice on what would be an effective explanation?"

[1202] "Please check the flow of conversations in real time while serving customers and let us know where we can improve."

[1203] "After you've finished explaining the product, please give us your feedback on what needs to be improved."

[1204] This is expected to enable efficient improvement of customer service skills and ultimately contribute to increased customer satisfaction.

[1205] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1206] Step 1: Data Capture

[1207] The smartphone device allows the user to perform a specific scenario (e.g., a new product presentation), and uses the device's camera and microphone to capture the user's voice and facial expression data in real time. This data is then sent raw to the next processing step.

[1208] Input: User's voice data, user's facial expression data

[1209] Output: Raw audio and facial expression data

[1210] Step 2: Send data

[1211] The captured voice and facial expression data is sent to the server using the smartphone's Requests library, where it is encoded and sent as an HTTP request.

[1212] Input: Raw audio and facial expression data

[1213] Output: The encoded data sent to the server.

[1214] Step 3: Convert audio data to text

[1215] The server converts the received voice data into text using automatic speech recognition (ASR), which converts the voice data into text data, making the voice content available as text.

[1216] Input: Encoded audio data

[1217] Output: Text data

[1218] Step 4: Feature extraction

[1219] The server extracts features from the text and audio data, such as speech stress, intonation, speaking rate, and text content, using speech analysis models and natural language processing (NLP) models.

[1220] Input: Text data, audio data

[1221] Output: Extracted features

[1222] Step 5: Assess your skill level

[1223] The server evaluates the user's skill level based on the extracted features, using a generative AI model to represent the user's customer service skills and presentation quality numerically and graphically.

[1224] Input: Extracted features

[1225] Output: User's skill level

[1226] Step 6: Provide a practice scenario

[1227] The server generates personalized practice scenarios based on the user's skill level, taking into account the user's weaknesses and areas for improvement, and suggests new scenarios.

[1228] Input: User's skill level

[1229] Output: A customized training scenario

[1230] Step 7: Capture user practice data

[1231] The user practices according to the provided practice scenario, while the smartphone again captures voice and facial expression data and sends them to the server. At the same time, the current practice status is monitored in real time.

[1232] Input: Voice data during practice, facial expression data during practice

[1233] Output: Practice data sent in real time

[1234] Step 8: Generate real-time feedback

[1235] The server analyzes the received practice data in real time and generates feedback for the user, including corrections to intonation, improvements to specific expressions, and adjustments to facial expressions.

[1236] Input: Voice data during practice, facial expression data during practice

[1237] Output: Real-time feedback

[1238] Step 9: Provide feedback

[1239] The smartphone device receives feedback from the server and presents it to the user in text, audio, visual, or other formats, allowing the user to use the feedback in their next practice.

[1240] Input: Real-time feedback

[1241] Output: Feedback presented to the user

[1242] This series of steps makes it possible to efficiently support users in improving their customer service skills.

[1243] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1244] This invention is a personalized dialogue simulator designed to improve skills in sales, presentations, meetings, and other areas. The system collects and analyzes the user's voice and facial expression data, provides customized practice scenarios, and provides real-time feedback. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it provides deeper feedback.

[1245] System program generation and processing details

[1246] 1. Data collection phase (server)

[1247] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[1248] The collected voice data is converted into text using automatic speech recognition (ASR) technology and saved as text information.

[1249] 2. Data preprocessing phase (server)

[1250] The server extracts features from the text and audio data, including parameters such as voice stress, intonation patterns, and speaking rate.

[1251] 3. Model training phase (server)

[1252] The server trains AI models to assess the user's skill level based on the extracted features, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[1253] The trained model is saved and used to generate specialized training scenarios for the user.

[1254] 4. Emotion engine integration (server)

[1255] The server incorporates a new emotion engine, which adds the ability to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation.

[1256] 5. User authentication phase (terminal)

[1257] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[1258] 6. User Interaction Phase (Terminal)

[1259] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[1260] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[1261] 7. Feedback Phase (Server and Device)

[1262] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[1263] The emotion engine recognizes the user's emotional state and assesses their stress level and motivation, which are then used to generate further feedback.

[1264] The server generates specific feedback based on the analysis, such as "You'll be more effective if you raise your voice tone here," "A quick smile here will make a better impression," or "You're under a lot of stress, so consider slowing down your practice."

[1265] The device displays the generated feedback to the user and provides specific instructions on how to improve for the next practice or presentation.

[1266] Specific examples

[1267] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the user's stress level and provides advice such as, "You may be too stressed and unable to concentrate on your practice. Take a break." The generated feedback is provided to the user via the device, and the user uses this information to carry out the next practice.

[1268] In this way, the system of the present invention significantly improves the quality of vocational training by providing practice scenarios optimized for each user and enabling real-time feedback. Furthermore, the introduction of an emotion engine provides detailed feedback according to the user's emotional state, enabling more practical and advanced skill improvement.

[1269] The processing flow will be explained below.

[1270] Step 1:

[1271] The server collects recorded audio data of sales proposals and presentations provided by companies, and stores this audio data as big data on the server.

[1272] Step 2:

[1273] The server converts the collected voice data into text using automatic speech recognition (ASR) technology, and the voice data is converted into text data and saved as text information.

[1274] Step 3:

[1275] The server extracts features from the text and audio data, calculating parameters such as voice stress, intonation patterns, and speaking rate.

[1276] Step 4:

[1277] The server trains AI models based on the preprocessed data to assess the user's skill level, including a speech analysis model, an intonation analysis model, and a facial expression recognition model.

[1278] Step 5:

[1279] The server incorporates a new emotion engine, which adds the ability to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation.

[1280] Step 6:

[1281] A user logs in to the system from a terminal and enters authentication information, which is then sent to the server for user authentication.

[1282] Step 7:

[1283] The terminal retrieves the individually tailored practice scenario from the server and presents it to the user, who then begins practicing based on the displayed scenario.

[1284] Step 8:

[1285] The device captures the user's voice and facial expression data in real time while they practice, and sends them to the server for analysis.

[1286] Step 9:

[1287] The server analyzes the received voice and facial expression data in real time, including recognizing voice stress, intonation changes, and facial expressions.

[1288] Step 10:

[1289] The server generates specific feedback based on the analysis, such as "It would be more effective if you raised your voice tone in this part" or "If you smile briefly here, you'll make a better impression."

[1290] Step 11:

[1291] The emotion engine recognizes the user's emotional state from their voice and facial expression data and evaluates their stress level and motivation. For example, if the user's voice tone is low and their facial expression is stiff, it may determine that the user is feeling stressed.

[1292] Step 12:

[1293] The server generates further feedback based on the emotion engine's evaluation, such as advice like, "You may be under a lot of stress and not be able to concentrate on your practice. Take a break."

[1294] Step 13:

[1295] The device displays the generated feedback to the user and provides specific instructions on how to improve in the next practice or presentation. Based on the feedback, the user can adjust their voice tone, intonation, and facial expressions.

[1296] Step 14:

[1297] The user receives the provided feedback and practices the next time. Based on the feedback, the user can adjust the tone, intonation, and facial expressions of their voice to maximize the effectiveness of their training.

[1298] Example 2

[1299] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1300] Conventional vocational training systems have difficulty providing feedback that fully considers the characteristics and emotional state of individual users. This makes it difficult to efficiently improve users' practical skills. In addition, real-time feedback generation is often delayed, making it difficult to provide immediate improvements.

[1301] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice data, means for converting the voice data into text, means for extracting features from the text data and voice data, means for evaluating the user's skill level based on the extracted features, means for recognizing the user's emotional state, means for presenting a practice scenario individually tailored to the user, means for capturing voice and facial expression data of the user during practice and transmitting the captured data to the server, means for analyzing the captured data in real time and generating feedback for the user, and means for presenting the generated feedback to the user. This makes it possible to provide a practice scenario individually optimized for the user and generate feedback instantly.

[1302] "Voice data" is digital data that records what the user has said.

[1303] "Text data" refers to character information generated from voice data using automatic voice recognition technology.

[1304] "Features" are parameters extracted from speech and facial expression data and used for analysis and model training. Specifically, they include speech stress, intonation patterns, speaking rate, etc.

[1305] The "means for assessing skill level" refers to an algorithm or model for assessing a user's ability or performance based on the extracted features.

[1306] The "means for recognizing emotional state" refers to an engine or technology that analyzes the user's voice data and facial expression data to grasp the user's emotional state, such as stress and motivation level.

[1307] The "means for presenting a practice scenario" is an interface or system for presenting practice content that is specialized for each individual user.

[1308] The "capturing means" refers to a device or software for collecting voice and facial expression data in real time while the user is practicing.

[1309] The "means for generating feedback" refers to an algorithm or system that analyzes the captured voice and facial expression data and presents areas for improvement to the user.

[1310] A "means for presenting feedback" is an interface or device that visually or audibly notifies the user of the generated feedback.

[1311] This invention is a personalized dialogue simulator designed to improve skills in sales, presentations, meetings, and other areas. The system collects and analyzes the user's voice and facial expression data, provides customized practice scenarios, and provides real-time feedback. Furthermore, by combining it with an emotion engine that recognizes the user's emotional state, it provides deeper feedback.

[1312] First, the server collects recorded audio data of sales proposals and presentations provided by companies. This audio data is stored on the server as big data. The collected audio data is converted into text using automatic speech recognition (ASR) technology and saved as text information. A variety of automatic speech recognition APIs can be used for the ASR technology used.

[1313] The server then extracts features from the text and audio data, such as voice stress, intonation patterns, and speaking rate. A speech signal processing library is used to extract the features, and a computer vision library is used to analyze facial expression data.

[1314] The server then trains AI models based on these feature data. The trained models include a speech analysis model, an intonation analysis model, and a facial expression recognition model. The deep learning frameworks used include commonly used deep learning platforms.

[1315] The server also incorporates an emotion engine to recognize the user's emotional state. The emotion engine analyzes voice and facial expression data to assess the user's stress level and motivation. An emotion recognition API is available for this emotion analysis.

[1316] The user logs in to the system from their device and enters their authentication information. The device then sends this information to the server for user authentication. If authentication is successful, the device retrieves an individually tailored practice scenario from the server and presents it to the user. The user then begins practicing based on the displayed scenario. The device captures the user's voice and facial expression data in real time while they practice and sends the data to the server.

[1317] The server analyzes the received voice and facial expression data in real time. Analysis includes voice stress, changes in intonation, and facial expression recognition. The emotion engine recognizes the user's emotional state and evaluates their stress level and motivation. This evaluation result is used to generate further feedback. Based on the analysis results, the server generates specific feedback, such as "You should raise the tone of your voice in this part," "A quick smile here will create a good impression," or "You are under high stress, so consider slowing down your practice pace." The generated feedback is displayed to the user via their device, providing specific instructions on how to improve in their next practice or presentation.

[1318] As a concrete example, consider a user giving a presentation on a new product. The user logs in to the system and starts a specialized practice scenario. The device captures the user's voice and facial expressions while giving the presentation and sends them to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the user's stress level and provides advice such as, "You may be too stressed and unable to concentrate on your practice. Take a break." The generated feedback is provided to the user via the device, and the user uses this information to carry out the next practice.

[1319] Examples of input prompts include, "I would like the system to give me specific advice on what to pay attention to when presenting a new product." and "Please capture my voice and facial expression data during the presentation and evaluate it in real time."

[1320] In this way, the system of the present invention significantly improves the quality of vocational training by providing practice scenarios optimized for each user and generating immediate feedback. Furthermore, the introduction of an emotion engine provides detailed feedback according to the user's emotional state, enabling more practical and advanced skill improvement.

[1321] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1322] Step 1: Data collection

[1323] The server collects recorded audio data of sales proposals and presentations provided by companies. The server obtains audio data files from companies via HTTPS and stores them in data storage such as Google Cloud Storage. The input is the audio data file, and the output is the audio data file stored in the storage. The server converts the audio data into text data using the Google Cloud Speech-to-Text API and stores this text data in a database. The input is the audio data file, and the output is text data.

[1324] Step 2: Data Preprocessing

[1325] The server extracts features from text data and audio data. Features include voice stress, intonation patterns, speaking rate, etc. The Librosa library is used to extract features from the audio data and store them in a database. The input is audio data, and the output is feature data. The OpenCV library is used to analyze facial expression data, and facial expression features are also stored in the database. The input is facial expression image data, and the output is facial expression feature data.

[1326] Step 3: Model training

[1327] The server trains an AI model based on the collected feature data. Deep learning frameworks used include TensorFlow and PyTorch. For example, a speech analysis model is trained using TensorFlow and the model is saved. The input is the feature data, and the output is the trained AI model. Similarly, intonation analysis models and facial expression recognition models are trained and saved.

[1328] Step 4: Incorporating the Emotion Engine

[1329] The server incorporates an emotion engine into the system, adding the ability to recognize the user's emotional state. The emotion engine uses Affectiva or the Microsoft Emotion API. The API is used to analyze the user's emotions, and the results are stored in a database. The input is voice data and facial expression data, and the output is emotional state data.

[1330] Step 5: User authentication

[1331] A user logs in to the system from a terminal and enters authentication information. The terminal sends this information to the server, which then authenticates the user. OAuth2.0 or JWT is used for authentication. The input is the user authentication information, and the output is the authentication result.

[1332] Step 6: User Interaction

[1333] The device retrieves an individually tailored practice scenario from the server and presents it to the user. The user begins practicing based on the scenario displayed on the device. The device captures voice and facial expression data in real time and sends it to the server. The input is the practice scenario and the user's voice and facial expression data, and the output is the captured data.

[1334] Step 7: Generate feedback

[1335] The server analyzes the received voice and facial expression data in real time. The emotion engine also joins in to analyze the emotional state. For example, it analyzes the strength of the voice, changes in intonation, and recognizes facial expressions, and generates specific feedback. Based on the analysis results, it generates feedback such as "You should raise the tone of your voice in this part" or "If you smile briefly here, it will make a good impression." The input is the captured voice and facial expression data, and the output is the feedback data. The generated feedback is displayed to the user via the terminal.

[1336] The above is the specific processing flow of the program of this system.

[1337] (Application example 2)

[1338] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1339] Improving skills in sales, customer service, and presentations requires a lot of time and effort through manual education and training, but it is difficult to provide individually optimized feedback. There is also a lack of a system for evaluating the effectiveness of training in real time and immediately proposing improvement measures. Furthermore, it is difficult to maintain motivation for practice because feedback cannot take into account the emotional state of the trainee. This creates the problem of insufficient progress in improving customer service and presentation skills.

[1340] The identification processing by the identification processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes means for collecting user voice information, means for converting the voice information into character data, means for extracting features from the character data and the voice information, means for evaluating the user's skill level based on the extracted features, means for presenting an individually tailored practice scenario to the user, means for capturing the user's voice and facial expression information during practice and transmitting it to the server, means for analyzing the captured information in real time and generating feedback for the user, means for recognizing the user's emotional state, and means for presenting the generated feedback to the user. This makes it possible to effectively improve the user's skills through individually optimized practice scenarios and real-time feedback, and furthermore, by providing detailed feedback according to the user's emotional state, it is possible to achieve training effectiveness and maintain motivation.

[1341] "Voice information" refers to digitally recorded audio data, including the user's speech, tone of voice, intonation, and the like.

[1342] "Text data" refers to data that has been converted into text format by analyzing voice information, and is a written representation of what the user has said.

[1343] "Features" refer to measurable characteristics of a user's speech, such as stress, intonation, and speech rate, extracted from audio information and text data.

[1344] The "skill level" is an index that indicates the technical proficiency of the user's speaking style and facial expressions, evaluated based on the extracted feature amounts.

[1345] A "practice scenario" is an individually specialized practice scenario provided by the system for the purpose of improving the user's skills, and its content is customized according to the user's skill level.

[1346] "Facial expression information" refers to data that digitally records a user's facial expressions, including facial movements and emotional expressions.

[1347] The "emotional state" refers to the user's emotional state, which is recognized based on the user's facial expression information and voice information, and includes, for example, stress level and motivation level.

[1348] "Feedback" is specific advice about areas for improvement and effective practice methods for the user, generated based on analyzed voice information, text data, facial expression information, and emotional state.

[1349] "Real-time analysis" means processing the user's voice information and facial expression information instantly and obtaining analysis results instantly.

[1350] The present invention is an individually optimized dialogue simulator aimed at improving customer service skills in brick-and-mortar stores. This system involves communication between a server and a terminal, and provides users with optimized practice scenarios and feedback. A specific embodiment of this system is described below.

[1351] Data collection phase (server)

[1352] The server collects user voice information, which is uttered when the user practices customer service skills in a physical store and is recorded in real time using a smartphone or smart glasses.

[1353] Data preprocessing phase (server)

[1354] The collected speech information is converted into text data using automatic speech recognition (ASR) technology, such as Google Cloud Speech to Text. The Librosa library is then used to extract features from the speech data, including speech stress, intonation patterns, and speaking rate.

[1355] Model training phase (server)

[1356] The server trains AI models to assess the user's skill level based on the extracted features. These models include speech analysis, intonation analysis, and facial expression recognition. The trained models are saved and used to generate personalized practice scenarios for the user.

[1357] Emotion engine integration (server)

[1358] The server incorporates an emotion engine that recognizes the user's emotional state and analyzes it. The emotion engine is built using a deep learning library such as TensorFlow and evaluates the user's emotional state based on their voice and facial expression information.

[1359] User authentication phase (terminal)

[1360] The user logs in to the system from a terminal and begins practicing. The terminal can be a smartphone or smart glasses, and has the ability to capture the user's voice and facial expression information.

[1361] User Interaction Phase (Terminal)

[1362] The device retrieves an individually tailored practice scenario from the server and presents it to the user. The user begins practicing based on the displayed scenario, capturing voice and facial expression information in real time during the practice process and sending it to the server.

[1363] Feedback Phase (Server and Device)

[1364] The server analyzes the captured voice and facial expression information in real time and generates feedback. This feedback includes suggestions for improving voice stress, intonation, and facial expressions, as well as advice based on the user's emotional state. For example, the server provides specific feedback to the user, such as "It would be more effective if you raised the tone of your voice in this part," or "If you smile briefly here, you will make a good impression." The generated feedback is presented to the user via their device and can be used for future practice or when serving customers in physical stores.

[1365] Specific examples

[1366] For example, imagine a staff member introducing a new product. The staff member logs in to the system and starts a specialized practice scenario. The terminal captures the voice and facial expression information as the staff member explains the product's features and sends it to the server in real time. The server analyzes this data and generates feedback such as, "You should emphasize your intonation here," or "It would be good to smile a little at this part." Furthermore, the emotion engine detects the staff member's stress level and provides advice such as, "You may be too stressed and not able to concentrate on your practice. Take a break." The generated feedback is provided to the staff member via the terminal, and they use it to carry out their next practice.

[1367] Prompt Sentence Examples

[1368] Please generate a new customer service scenario. This scenario should be about a female customer in her twenties asking about a new product. The scenario should include the conversation with the customer, the information provided by the staff, and the features of the recommended product.

[1369] As described above, the system of the present invention supports the improvement of customer service skills in brick-and-mortar stores by providing practice scenarios optimized for each user and enabling real-time feedback. Furthermore, the introduction of an emotion engine allows for more detailed feedback based on the user's emotional state, enabling more effective training.

[1370] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1371] Step 1:

[1372] The server collects the user's voice information. This voice information is spoken by the user while practicing customer service skills and is recorded in real time using a device (smartphone or smart glasses). The input is the user's voice, and the output is digital voice data.

[1373] Step 2:

[1374] The server converts the collected voice information into text data using automatic speech recognition (ASR) technology. At this stage, the input is voice data and the output is text data. Specifically, an ASR service such as Google Cloud Speech to Text is used.

[1375] Step 3:

[1376] The server extracts features from text data and speech information. It uses the Librosa library to calculate features such as speech stress, intonation patterns, and speaking rate. The input is speech data and text data, and the output is a set of features.

[1377] Step 4:

[1378] The server evaluates the user's skill level based on the extracted features. This evaluation uses pre-trained AI models (voice analysis model, intonation analysis model, and facial expression recognition model). The input is the features, and the output is the skill level evaluation result.

[1379] Step 5:

[1380] The server generates a practice scenario specifically tailored for each user and sends it to the device. The scenario is customized based on the user's skill level assessment results. The input is the skill level assessment results, and the output is the practice scenario. The generative AI model generates the scenario according to the prompt sentence.

[1381] Step 6:

[1382] The terminal presents a practice scenario to the user, who then practices according to the scenario. The input is the practice scenario, and the output is the content of the practice performed by the user. Specifically, the scenario is displayed on the screen or guided by voice.

[1383] Step 7:

[1384] The device captures the user's voice and facial expression information while they practice and sends it to the server in real time. The input is the user's voice and facial expression information, and the output is the captured data. Specifically, the device's camera and microphone are always active.

[1385] Step 8:

[1386] The server analyzes the captured speech and facial expression information in real time and generates feedback for the user, including suggestions for improving intonation, vocal stress, facial expression, and emotional state. The input is the captured speech and facial expression information, and the output is the feedback.

[1387] Step 9:

[1388] The server uses an emotion engine to recognize the user's emotional state and reflects that information in the feedback. The input is captured voice and facial expression information, and the output is the evaluation result of the emotional state. Specifically, emotion analysis is performed using TensorFlow.

[1389] Step 10:

[1390] The terminal presents the generated feedback to the user. The input is the feedback, and the output is improvements or advice that is displayed or announced to the user by voice. Specific operations include displaying the feedback as text on the screen or announcing it by voice.

[1391] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1392] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1393] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1394] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1395] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1396] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1397] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1398] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1399] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1400] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1401] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1402] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1403] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1404] 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.

[1405] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1406] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1407] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1408] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1409] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1410] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1411] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1412] The following is further disclosed regarding the above embodiment.

[1413] (Claim 1)

[1414] means for collecting user voice data;

[1415] means for converting the voice data into text;

[1416] means for extracting features from the text data and the voice data;

[1417] means for evaluating a skill level of a user based on the extracted feature amount;

[1418] means for presenting a personalized practice scenario to a user;

[1419] A means for capturing voice and facial expression data of a user during practice and transmitting the data to a server;

[1420] means for analyzing the captured data in real time and generating feedback to a user;

[1421] means for presenting the generated feedback to a user;

[1422] A system including:

[1423] (Claim 2)

[1424] 2. The system of claim 1, wherein the feedback generating means generates the feedback by analyzing the user's intonation, vocal stress, and speaking rate.

[1425] (Claim 3)

[1426] 2. The system according to claim 1, wherein the feedback generating means generates feedback regarding improvement of facial expressions by analyzing facial expression data of the user.

[1427] "Example 1"

[1428] (Claim 1)

[1429] means for collecting user voice data;

[1430] means for converting the voice data into text;

[1431] means for extracting features from the text data and the voice data;

[1432] means for training a generative AI model that assesses a user's skill level based on the extracted features;

[1433] means for presenting a personalized practice scenario to a user;

[1434] A means for capturing voice and facial expression data of a user during practice and transmitting the data to a server;

[1435] means for analyzing the captured data in real time and generating feedback to a user;

[1436] means for presenting the generated feedback to a user;

[1437] A system including:

[1438] (Claim 2)

[1439] 2. The system of claim 1, wherein the feedback generating means generates the feedback by analyzing the user's intonation, vocal stress, and speaking rate.

[1440] (Claim 3)

[1441] 2. The system according to claim 1, wherein the feedback generating means generates feedback regarding improvement of facial expressions by analyzing facial expression data of the user.

[1442] "Application Example 1"

[1443] (Claim 1)

[1444] means for collecting user voice data;

[1445] means for converting the voice data into text;

[1446] means for extracting features from the text data and the voice data;

[1447] means for evaluating a skill level of a user based on the extracted feature amount;

[1448] means for presenting a personalized practice scenario to a user;

[1449] A means for capturing voice and facial expression data of a user during practice and transmitting the data to a server;

[1450] means for transmitting voice and facial expression data during the customer service to a server in real time;

[1451] means for analyzing the captured data in real time and generating feedback to a user;

[1452] means for presenting the generated feedback to a user;

[1453] A means for providing a user with practice scenarios for improving customer service skills;

[1454] A system including:

[1455] (Claim 2)

[1456] 2. The system of claim 1, wherein the feedback generating means generates the feedback by analyzing the user's intonation, vocal stress, and speaking rate.

[1457] (Claim 3)

[1458] 2. The system according to claim 1, wherein the feedback generating means generates feedback regarding improvement of facial expressions by analyzing facial expression data of the user.

[1459] "Example 2: Combining Emotion Engines"

[1460] (Claim 1)

[1461] means for collecting user voice data;

[1462] means for converting the voice data into text;

[1463] means for extracting features from the text data and the voice data;

[1464] means for evaluating a skill level of a user based on the extracted feature amount;

[1465] means for recognizing the emotional state of a user;

[1466] means for presenting a personalized practice scenario to a user;

[1467] A means for capturing voice and facial expression data of a user during practice and transmitting the data to a server;

[1468] means for analyzing the captured data in real time and generating feedback to a user;

[1469] means for presenting the generated feedback to a user;

[1470] A system including:

[1471] (Claim 2)

[1472] 2. The system of claim 1, wherein the feedback generating means generates the feedback by analyzing the user's intonation, vocal stress, and speaking rate.

[1473] (Claim 3)

[1474] 2. The system of claim 1, wherein the feedback generating means generates feedback regarding facial expression and emotional improvement by analyzing the user's facial expression data and emotional state.

[1475] "Application example 2 when combining emotion engines"

[1476] (Claim 1)

[1477] means for collecting user voice information;

[1478] means for converting the voice information into character data;

[1479] means for extracting features from the character data and voice information;

[1480] means for evaluating a skill level of a user based on the extracted feature amount;

[1481] means for presenting a personalized practice scenario to a user;

[1482] A means for capturing voice and facial expression information of a user during practice and transmitting the captured information to a server;

[1483] means for analyzing the captured information in real time and generating feedback to a user;

[1484] means for recognizing the emotional state of a user;

[1485] means for presenting the generated feedback to a user;

[1486] A system including:

[1487] (Claim 2)

[1488] 2. The system of claim 1, wherein the feedback generating means generates the feedback by analyzing the user's intonation, vocal stress, and speaking rate.

[1489] (Claim 3)

[1490] 2. The system according to claim 1, wherein the feedback generating means generates feedback regarding improvement of facial expressions by analyzing facial expression information of the user.

[1491] (Claim 4)

[1492] 2. The system according to claim 1, wherein the feedback generating means generates feedback regarding stress level and motivation by analyzing the user's emotional state. [Explanation of symbols]

[1493] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for collecting user voice data; means for converting the voice data into text; means for extracting features from the text data and the voice data; means for evaluating a skill level of a user based on the extracted feature amount; means for presenting a personalized practice scenario to a user; A means for capturing voice and facial expression data of a user during practice and transmitting the data to a server; means for analyzing the captured data in real time and generating feedback to a user; means for presenting the generated feedback to a user; A system including:

2. 2. The system of claim 1, wherein the feedback generating means generates the feedback by analyzing the user's intonation, vocal stress, and speaking rate.

3. 2. The system according to claim 1, wherein the feedback generating means generates feedback regarding improvement of the facial expression by analyzing the facial expression data of the user.

Images