system

Abstract

We provide the system. [Solution] An image generation means for detecting the facial expressions of visitors, An emotion classification means for classifying the emotions of visitors based on the aforementioned facial expressions, A means of acquiring audio for obtaining the voice of a visitor, A voice emotion estimation means for estimating emotions based on the aforementioned voice, A data analysis means for integrating and analyzing the aforementioned emotional data based on facial expressions and voice, A system including a report generation means for generating output information using the analyzed emotion data.

Description

Technical Field

[0001] The technology disclosed herein relates to a system.

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the food service industry, in order to improve customer satisfaction, it is required to accurately and quickly understand and respond to the emotional state of individual customers. However, currently, it is difficult to directly grasp the emotions of customers, and subjective judgments are often relied upon. Therefore, it is difficult to accurately capture the instantaneous emotional changes of customers in real time and efficiently implement appropriate service improvements based on feedback.

Means for Solving the Problems

[0005] This invention provides an emotion classification means that uses an image generation means to detect a visitor's facial expression and accurately classify the visitor's emotions based on it. It also includes a voice acquisition means to acquire the visitor's voice data and a voice emotion estimation means to estimate their emotions. Furthermore, by including a report generation means that comprehensively analyzes this emotion data using a data analysis means and generates output information based on the results, the invention provides a system that allows service personnel to obtain accurate and rapid feedback when needed. This enables real-time understanding of customer emotions and appropriate service improvements.

[0006] "Image generation means" refers to technology that has the function of acquiring and processing image or video data in order to detect the facial expressions of visitors.

[0007] "Emotion classification means" refers to a technology that analyzes a visitor's facial expression based on acquired image data and classifies it into a specific emotional category.

[0008] "Voice acquisition means" refers to technology that has the function of acquiring voice data of visitors and processing that data into an analyzable format.

[0009] A "voice emotion estimation means" is a technology that has the function of extracting characteristics from acquired voice data and estimating emotions based on those characteristics.

[0010] "Data analysis means" refers to a technology that has the function of integrating and analyzing emotional information obtained from image data and audio data.

[0011] A "report generation method" is a technology that has the function of generating output information based on analyzed emotional information and providing it as a report. [Brief explanation of the drawing]

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

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

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

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

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

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

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

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

[0020] [First Embodiment]

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

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

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

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

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

[0026] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

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

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

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

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

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

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

[0033] This invention provides a system for service industries such as restaurants that can grasp customer emotions in real time and improve the quality of service. This system mainly consists of a server, terminals, and users.

[0034] The server receives video data from cameras to detect visitors' facial expressions. Cameras are installed in multiple locations within the store and can capture visitors' facial expressions. The server recognizes facial regions from the video data and applies an expression analysis algorithm to classify the visitor's emotions. This classification can be divided into categories such as joy, surprise, and dissatisfaction.

[0035] Furthermore, the server receives audio data from microphones installed within the store. The terminals are responsible for transmitting the audio data to the server. The server analyzes the audio data, extracts specific voice features, and estimates the emotion. This analysis is based on factors such as the pitch, speed, and volume of the voice.

[0036] The server integrates emotional data from video and audio and records it in a database. This ensures that customer emotional data is stored in a time-synchronized manner. Based on this data, the server performs data analysis and generates reports.

[0037] The generated report is sent to the administrator's terminal, where store managers and staff can review it. This report includes customer sentiment trends, frequent patterns, and recommendations for service improvement.

[0038] As a concrete example, suppose a visitor is eating a meal, and the camera captures their facial expression, and the server detects an emotion of "dissatisfaction." At the same time, the server analyzes the audio picked up by the microphone and detects a statement such as "the food is cold." Based on this information, the server generates a report titled "Dissatisfaction regarding the food temperature" and notifies the terminal. The user (staff) can then use this feedback to quickly take measures such as reheating the food. In this way, this system helps to improve customer satisfaction.

[0039] The following describes the processing flow.

[0040] Step 1:

[0041] The server acquires video data in real time from cameras installed in the store. The video data is transmitted from multiple cameras installed at each table.

[0042] Step 2:

[0043] The server identifies the visitor's face from the acquired video data using a face region detection algorithm. An expression analysis algorithm is then applied to the identified face region to classify it into a specific emotion category.

[0044] Step 3:

[0045] The terminal transmits audio data collected within the store to a server. This audio data is recorded by microphones installed at each table.

[0046] Step 4:

[0047] The server processes the transmitted audio data through an audio preprocessing engine to remove noise and convert it into a clear signal. Next, it uses an audio processing algorithm to extract audio features and perform emotion estimation based on the audio data.

[0048] Step 5:

[0049] The server integrates emotion categories from images and emotion estimation results from audio, and records the visitor's overall emotional state in a database.

[0050] Step 6:

[0051] The server analyzes accumulated emotional data to identify customer emotional trends and automatically generates reports based on this analysis.

[0052] Step 7:

[0053] The terminal displays the generated reports to store managers and staff. Users (staff) can use this information to implement operational improvement measures in real time.

[0054] (Example 1)

[0055] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0056] In today's service industry, improving customer satisfaction requires understanding customers' emotional states in real time and responding appropriately. However, conventional methods have made it difficult to accurately analyze emotions from customers' facial expressions and voices and effectively provide feedback. This invention solves this problem and provides a means to quickly and accurately detect customer emotions and use that information to improve services.

[0057] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0058] In this invention, the server includes image processing means for detecting a visitor's facial expression, data classification means for classifying the visitor's emotions based on information extracted from the facial expression, and voice processing means for acquiring and processing the visitor's voice. This makes it possible to integrate and analyze emotion data extracted from the customer's facial expression and voice, generate appropriate reports in real time, and improve service.

[0059] "Image processing means" refers to a technology for detecting a visitor's facial expression and extracting facial features from that expression.

[0060] "Data classification means" refers to a method of classifying visitors' emotions into multiple categories based on information obtained through image processing means.

[0061] "Voice processing means" refers to a method of acquiring the voice of a visitor and performing preprocessing on that voice signal.

[0062] "Acoustic analysis means" refers to techniques for estimating a visitor's emotions using the characteristics of their voice.

[0063] "Information analysis means" refers to a process that integrates data obtained from image processing means and audio processing means to comprehensively analyze customer emotions.

[0064] "Information generation means" refers to a function that uses analyzed sentiment data to create report information and provides feedback for service improvement.

[0065] This invention provides a system that analyzes customer emotions in real time in the service industry and improves service quality. This system mainly consists of a server, terminals, and users.

[0066] The server receives video data of visitors from cameras installed within the store. When identifying facial regions from this video data, the server uses an image processing library such as OpenCV. In addition, to analyze facial expressions and classify emotions, it performs emotion classification using machine learning models such as TENSORFLOW®.

[0067] The terminal receives audio data from microphones in the store and sends it to the server. The server analyzes the audio data using an audio processing library such as LibROSA and estimates emotions from the audio features. This analysis is based on the pitch, speed, and volume of the speech.

[0068] Users (e.g., service staff) can view reports generated by the server on their administrator terminals, enabling immediate responses based on the customer's emotional state. This can improve customer satisfaction.

[0069] As a concrete example, consider a situation where a visitor is eating a meal. In this case, the camera captures their facial expression, and the server detects an emotion of "dissatisfaction." Then, the server analyzes the audio captured through the microphone and determines that the visitor said, "The food is cold." As a result, the server generates a report of "dissatisfaction regarding the temperature of the food" and notifies the terminal. Consequently, the user can rationally take action to reheat the food.

[0070] (Example of prompts for a generative AI model)

[0071] "Please describe a system that uses cameras and microphones installed in restaurants to analyze the facial expressions and voices of customers and understand their emotions in real time. Include specific hardware and software names, usage instructions, and examples."

[0072] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0073] Step 1:

[0074] The server receives real-time video data from cameras installed within the store. Using this video data as input, it extracts face regions using the OpenCV library. This process yields output of face coordinates, which determine the position and size of the faces. Specifically, this involves identifying faces frame by frame and preparing the information necessary for subsequent processing.

[0075] Step 2:

[0076] The server analyzes facial expressions using machine learning models such as TensorFlow, based on facial region information. Using the obtained facial coordinates as input data, it classifies expressions into emotional categories such as joy, surprise, and dissatisfaction. The output of this process is a numerical emotion score. Specifically, for example, a score of 0.8 would represent "joy."

[0077] Step 3:

[0078] The device transmits audio data acquired through the microphone to the server. The transmitted audio data is received as input, and the server uses the LibROSA library to extract audio features. The output obtained here consists of numerical acoustic features such as pitch, velocity, and volume. Specifically, if the audio has a higher pitch than normal, that value is measured.

[0079] Step 4:

[0080] The server performs analysis to estimate emotions from speech based on the extracted acoustic features. Using the acoustic features analyzed by LibROSA as input, it uses a machine learning algorithm to estimate emotions. The output of this process is a score for each emotion category. Specifically, emotions such as "excitement" are scored and output based on factors like pitch and speed.

[0081] Step 5:

[0082] The server integrates emotional data from video and audio. It takes the emotional scores from each analysis step as input and combines them to create consistent customer emotional data. This output is a comprehensive emotional profile showing how the customer felt in the store. Specifically, it yields integrated emotional scores such as "Joy 0.8, Excitement 0.6".

[0083] Step 6:

[0084] The server analyzes the integrated emotional profiles and generates reports. It takes the emotional profiles obtained during the integration process as input, analyzes the output, and creates reports that include recommendations for service improvement. Specifically, it aggregates the emotions customers expressed during specific events or with specific products and presents them as areas for improvement in the store.

[0085] (Application Example 1)

[0086] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0087] In today's service industry, improving customer satisfaction requires accurately understanding customer emotions and responding promptly and appropriately. However, traditional systems have struggled to grasp customer intentions and emotions in real time and respond immediately, resulting in limitations in improving customer satisfaction. Furthermore, employees have limitations in their ability to directly read customers' emotions, which hinders effective customer service.

[0088] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0089] In this invention, the server includes an image generation means for detecting the visitor's facial expression, a notification means for notifying an information display device of the emotional data in real time to enable immediate response, and a display means for displaying the analyzed emotional data to facilitate service based on the visitor's state. This allows employees to grasp the customer's emotions in real time and immediately use that information to improve service, thereby increasing customer satisfaction.

[0090] The "image generation means" is a method for detecting a visitor's facial expression using a camera and generating image data in real time.

[0091] An "emotion classification method" is a means of classifying a visitor's emotions into pre-defined categories based on acquired facial expression data.

[0092] "Voice acquisition means" refers to a means for acquiring the voice of a visitor and converting that data into an analyzable format.

[0093] A "voice emotion estimation means" is a voice analysis means used to identify the emotions of a visitor based on acquired voice data.

[0094] "Data analysis means" refers to a method for integrating and comprehensively analyzing facial expression data and voice data.

[0095] A "report generation method" is a means of generating output information by utilizing analyzed sentiment data.

[0096] A "notification means" is a means of transmitting analyzed emotion data to an information display device in real time and immediately notifying the user of related information.

[0097] A "display means" is a means of visually displaying information so that users can perform appropriate services based on emotional data.

[0098] The system that realizes this application consists of the following elements: The server plays a central role in the customer service support system using smart glasses. First, the server receives video data from cameras installed in the store and uses this data to recognize the facial expressions of visitors. OpenCV is used for image processing, and a face recognition algorithm is applied to identify individual face regions. Based on this face data, emotions are classified using TensorFlow.

[0099] Next, the server receives audio data from the microphone and uses PyDub and Sphinx for speech analysis. This analysis extracts features such as pitch, speed, and volume from the speech and performs speech emotion estimation. The resulting emotion data is then integrated by the server and sent in real time to the administrator terminal via Firebase.

[0100] On the other hand, the staff, who are the users, wear smart glasses and can check customer emotional information in real time through the display device. This is because emotional data is displayed on the glasses via a notification system. For example, when a customer places an order with a staff member wearing smart glasses, if the customer shows an anxious expression regarding the order, this information will be displayed on the glasses' screen as, "The customer appears anxious. The order needs to be confirmed." Based on this information, the user can quickly take action to alleviate the customer's anxiety.

[0101] An example of a prompt using the generative AI model might be, "Please explain how to generate customer sentiment data that integrates customer facial expressions and voice, and can be referenced by employees in real time." In this way, the system improves customer interaction and enhances the quality of service.

[0102] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0103] Step 1:

[0104] The server receives video data in real time from cameras installed in the store. It processes the video data as input using OpenCV, and identifies individual face regions using a face recognition algorithm. This process yields face position information and facial expression data as output.

[0105] Step 2:

[0106] The server uses TensorFlow to classify the facial expression data extracted in the previous step into emotion categories. The input data is facial expression data, and the server analyzes this data to output emotion tags such as "joy" and "dissatisfaction." Based on this data, the server is ready to provide information to the staff.

[0107] Step 3:

[0108] The server receives audio data from microphones in the store. The input audio is preprocessed with PyDub, and speech analysis is performed using Sphinx. Features such as pitch and volume are extracted, and emotion estimation is performed to obtain emotion-related speech features as output.

[0109] Step 4:

[0110] The server integrates facial expression data and voice data. Here, both sets of data are analyzed and the resulting emotional data is structured using Firebase. The output is emotional data formatted for use in decision-making.

[0111] Step 5:

[0112] The server notifies staff members in real time of integrated emotional data via smart glasses. This notification system allows staff to display customer emotional information (the input data) on the smart glasses, enabling immediate responses. The display includes specific customer service advice.

[0113] Step 6:

[0114] Staff members, acting as users, select actions based on customer emotion information displayed on the smart glasses' screen. During interactions with customers, they choose actions to improve service based on the emotion information. This ensures that emotion data, as input, is reflected in practice, leading to improved customer satisfaction.

[0115] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0116] This invention provides a system for accurately and in real time understanding customer emotions in the service industry and improving the quality of service. This system mainly consists of a server, terminals, users, and an emotion engine.

[0117] The server receives video data in real time from cameras installed in the store and applies video processing algorithms to detect visitors' facial expressions. Based on these detected expressions, an emotion classification system categorizes the visitor's emotions. This classification is divided into categories such as joy, surprise, sadness, and dissatisfaction, and the customer's instantaneous emotions are also recorded.

[0118] Simultaneously, the server processes the audio data transmitted from the terminal. The terminal collects audio data through microphones installed in the store and transmits it to the server. The server performs preprocessing such as noise reduction and audio normalization, and estimates the emotion from the audio using an audio emotion estimation method.

[0119] Furthermore, an emotion engine is integrated into the system, enabling multifaceted emotion recognition that includes user behavior data and biometric signals. The emotion engine uses machine learning models to analyze various collected data and recognize the customer's emotional state with high accuracy.

[0120] The server integrates and analyzes facial expression data, voice data, and data from the emotion engine. Based on this analysis, the server generates a report in real time and sends it to the terminal. This report is viewable by administrators and staff and contains customer emotion trends and recommended actions for service improvement.

[0121] For example, if a visitor expresses dissatisfaction during a meal, the camera captures their facial expression, and the server classifies it as "dissatisfaction." Simultaneously, if the word "slow" is extracted from the audio data, the emotion engine, processing this information, recognizes that the user is dissatisfied with the slow service. This information is immediately generated as a report and notified to the staff's terminal. Based on this, the user (staff) can quickly implement countermeasures to improve customer satisfaction. In this way, this system enables real-time emotion recognition and on-the-spot service improvement, resulting in a better customer experience.

[0122] The following describes the processing flow.

[0123] Step 1:

[0124] The server receives video data in real time from cameras installed within the store. The cameras are positioned to capture each table where customers are seated.

[0125] Step 2:

[0126] The server applies a face region detection algorithm to the received video data to identify the faces of individual visitors. A deep learning model is then used to analyze the facial expressions of the identified face regions and classify the emotions.

[0127] Step 3:

[0128] The terminal collects the audio of visitors' conversations through microphones in the store and simultaneously transmits that audio data to a server.

[0129] Step 4:

[0130] The server performs noise reduction and normalization on the received audio data. Next, it applies an audio emotion estimation algorithm to estimate the emotion based on the audio.

[0131] Step 5:

[0132] The server integrates emotional data obtained from facial expressions and emotional data obtained from voice. It then activates an emotion engine and uses a machine learning-based model to accurately recognize the user's emotions from the integrated data.

[0133] Step 6:

[0134] The server analyzes the recognized emotion data to understand customer emotion trends and generates a report based on the analysis results.

[0135] Step 7:

[0136] The terminal receives reports sent from the server and displays them to administrators and staff. Based on these reports, users (staff) take appropriate actions in real time to improve customer satisfaction.

[0137] (Example 2)

[0138] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0139] In the service industry, accurately understanding customer emotions in real time and improving service quality is essential. However, conventional systems have been insufficient in analyzing data obtained from video and audio, making it difficult to understand customer emotions from multiple perspectives. Furthermore, it has been difficult to appropriately utilize the analysis results to improve customer service.

[0140] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0141] In this invention, the server includes video processing means for detecting a visitor's facial expression, emotion classification means for classifying the visitor's emotions based on the facial expression, and acoustic data collection means for acquiring the visitor's voice. This enables multifaceted emotional recognition of the customer.

[0142] "Image processing means" refers to technology for detecting visitors' facial expressions in real time and analyzing the image data.

[0143] "Emotion classification means" refers to a technology that classifies and categorizes a visitor's emotions based on the characteristics of their facial expressions.

[0144] "Acoustic data collection means" refers to technology for collecting sounds generated within a store and pre-processing them into a format suitable for analysis.

[0145] "Voice emotion estimation means" refers to a technology for estimating a visitor's emotions from collected voice data.

[0146] "Emotion recognition means" refers to a technology that analyzes multiple data points, including user behavior data and biosignals, from various angles to recognize visitors' emotions with high accuracy.

[0147] "Data analysis means" refers to technology for integrating the results of facial expression, voice, and emotion recognition and performing detailed analysis.

[0148] "Information generation means" refers to technology for generating output information, including measures to improve customer service, based on analyzed emotional data.

[0149] This invention is configured as a system for understanding customer emotions in the service industry in real time and improving the quality of service. This system consists of multiple elements, including a server, terminals, users, and an emotion engine.

[0150] The server receives real-time video data from high-resolution cameras installed within the store. The server uses video processing algorithms to detect visitors' facial expressions and analyzes this data using emotion classification tools. This allows the server to classify visitors' emotions into categories such as "joy," "surprise," "sadness," and "dissatisfaction."

[0151] The terminal collects audio data through multiple microphones installed within the store and transmits it to a server. The server performs noise reduction and speech normalization on the audio data and estimates the customer's emotions using speech emotion estimation tools. These processes can utilize common open-source speech processing libraries or proprietary speech analysis tools.

[0152] Furthermore, the system incorporates an emotion engine that collects and analyzes multifaceted data, including behavioral data and biosignals. The emotion engine utilizes a generative AI model and machine learning to accurately recognize customer emotions. This allows the server to analyze multiple integrated emotional data sets and generate reports based on the results.

[0153] Users (staff) receive reports notified to their terminals and use the analysis results to improve customer service. For example, if a visitor expresses dissatisfaction in the store, the server analyzes their facial expressions and voice data and creates a report so that staff can respond immediately.

[0154] The following is an example of a specific example and prompt: "If a customer in a restaurant expresses dissatisfaction during their meal, explain how you would analyze their emotions in real time and communicate that information to staff to address the issue."

[0155] The advantage of this system lies in its ability to enable real-time emotion recognition, allowing for quick and effective implementation of countermeasures to improve the customer experience.

[0156] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0157] Step 1:

[0158] The terminal uses cameras installed in the store to acquire real-time video data of visitors. This video data becomes the input to the system. The terminal sends this data directly to the server to prepare for subsequent processing. The actions performed here include camera adjustment and motion detection.

[0159] Step 2:

[0160] The server applies a face detection algorithm to the received video data. This identifies face regions from each frame. The output of this process is the coordinate data of each face region. By identifying the position of the faces, the data necessary for subsequent facial expression analysis is prepared.

[0161] Step 3:

[0162] The server applies an expression recognition algorithm to the data after face detection is complete, classifying emotions. In this process, emotions such as joy, surprise, sadness, and dissatisfaction are determined from facial feature points. The output is a label for the emotion category.

[0163] Step 4:

[0164] The terminal collects customer voice data using microphones installed in the store. This voice data is input to a server, where it undergoes pre-processing such as noise reduction. Since the collected data includes conversation and background noise, it is adjusted to extract important voice components.

[0165] Step 5:

[0166] The server receives denoised audio data and applies an audio emotion estimation algorithm to it. Based on the features extracted from the audio, it estimates the customer's emotions. The output is an emotion label from the audio, ready to be integrated with facial expression data.

[0167] Step 6:

[0168] The server integrates emotional information obtained from both facial expression and voice data into an emotion engine. This emotion engine uses a generative AI model and machine learning to perform comprehensive emotional analysis. The analysis results of multiple emotional data are output and transformed into a form that is meaningful to the user.

[0169] Step 7:

[0170] The server generates a report based on the analyzed sentiment data. The output includes customer sentiment trends and suggestions for service improvement. This report is sent to terminals, allowing users (staff) to understand it in real time and use it to improve their responses.

[0171] (Application Example 2)

[0172] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0173] In traditional service industries, it is difficult to instantly grasp and respond to customer emotions, limiting the potential for improving service quality. There is a need for technology that can analyze customer emotions in real time and enable service providers to respond appropriately on the spot.

[0174] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0175] In this invention, the server includes a video generation means for detecting a visitor's facial expression, an emotion classification means for classifying the visitor's emotions based on the facial expression, and an audio acquisition means for acquiring the visitor's sound. This allows service staff to recognize customers' emotions in real time and immediately improve their service.

[0176] The "image generation means" is a mechanism that acquires and processes the video data necessary to detect the facial expressions of visitors.

[0177] An "emotion classification system" is a mechanism that classifies a visitor's emotions based on acquired facial expression data.

[0178] "Sound acquisition means" refers to a mechanism for acquiring sound data of visitors.

[0179] An "acoustic emotion estimation means" is a mechanism that estimates emotions based on acoustic data.

[0180] "Information analysis means" refers to a mechanism that integrates and analyzes emotional data based on facial expressions and sounds.

[0181] An "information generation means" is a mechanism that generates output information using analyzed emotion data.

[0182] A "wearable display device" is a device equipped with a display device for presenting output information to customers.

[0183] This invention is a system that grasps customer emotions in real time in physical stores and improves the quality of service. The server processes video data acquired from cameras installed in the store using video generation means for detecting visitors' facial expressions and performs facial recognition. Furthermore, it collects acoustic data from microphones installed in the store using acoustic acquisition means for acquiring acoustic data of visitors and estimates acoustic emotions.

[0184] The server uses information analysis tools to comprehensively analyze this data and recognize the customer's emotional state with high accuracy. From the analyzed emotional data, an information generation tool generates output information, which is then presented to the service provider on the spot via a wearable display device. This information provides crucial clues for the service provider to take quick action.

[0185] Specific hardware includes wearable display devices such as smart glasses. Software used includes OpenCV for video processing, librosa for audio processing, and scikit-learn for machine learning models. This enables real-time customer emotion recognition and rapid service improvement. A concrete application of the invention is that, for example, if a customer shows dissatisfaction while waiting, their facial expression and sound can be used to recognize their emotion as "dissatisfaction," allowing staff to respond quickly and improve customer satisfaction.

[0186] An example of a specific prompt for the generating AI model is, "Based on the visitor's facial expressions and audio data, please list the most suitable recommended actions to improve customer service on the spot." This allows service staff to quickly implement appropriate countermeasures.

[0187] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0188] Step 1:

[0189] The server acquires video data from cameras installed within the store. The input is real-time video data, and the output is image data including the facial regions of visitors. The video generation method uses OpenCV to perform face recognition and extracts facial regions of particular interest.

[0190] Step 2:

[0191] The server acquires acoustic data from microphones installed within the store. The input is acoustic data including ambient noise, and the output is pre-processed acoustic data. The acoustic acquisition method uses librosa to remove noise and normalize the acoustic signal.

[0192] Step 3:

[0193] The server classifies emotions based on facial expression data. The input is image data of the face region, and the output is the classified emotion category (e.g., joy, surprise, dissatisfaction). The emotion classification means analyzes facial landmarks and estimates emotions using a machine learning model.

[0194] Step 4:

[0195] The server estimates acoustic emotion based on acoustic data. The input is normalized acoustic data, and the output is the estimated emotion category. The acoustic emotion estimation means performs emotion estimation based on voice tone and phoneme information.

[0196] Step 5:

[0197] The server integrates and analyzes emotional information obtained from facial expression data and acoustic data. The input is multiple emotional data, and the output is the overall emotional state of the customer. The information analysis means integrates different emotional indicators with weights to generate an overall picture.

[0198] Step 6:

[0199] The server generates output information based on the analysis results. The input is the overall emotional state, and the output is a report containing recommended actions for improving customer service. The information generation means uses a generation AI model to create a list based on prompt sentences.

[0200] Step 7:

[0201] The user receives output information generated through a wearable display device. The input is a real-time report, and the output manifests as a rapid action taken by the service provider. Display devices such as smart glasses visually notify the service provider of the emotional state.

[0202] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0203] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0204] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0205] [Second Embodiment]

[0206] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0207] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0208] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0209] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0210] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0211] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0212] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0213] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0214] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0215] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0216] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0217] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0218] This invention provides a system for service industries such as restaurants that can grasp customer emotions in real time and improve the quality of service. This system mainly consists of a server, terminals, and users.

[0219] The server receives video data from cameras to detect visitors' facial expressions. Cameras are installed in multiple locations within the store and can capture visitors' facial expressions. The server recognizes facial regions from the video data and applies an expression analysis algorithm to classify the visitor's emotions. This classification can be divided into categories such as joy, surprise, and dissatisfaction.

[0220] Furthermore, the server receives audio data from microphones installed within the store. The terminals are responsible for transmitting the audio data to the server. The server analyzes the audio data, extracts specific voice features, and estimates the emotion. This analysis is based on factors such as the pitch, speed, and volume of the voice.

[0221] The server integrates emotional data from video and audio and records it in a database. This ensures that customer emotional data is stored in a time-synchronized manner. Based on this data, the server performs data analysis and generates reports.

[0222] The generated report is sent to the administrator's terminal, where store managers and staff can review it. This report includes customer sentiment trends, frequent patterns, and recommendations for service improvement.

[0223] As a concrete example, suppose a visitor is eating a meal, and the camera captures their facial expression, and the server detects an emotion of "dissatisfaction." At the same time, the server analyzes the audio picked up by the microphone and detects a statement such as "the food is cold." Based on this information, the server generates a report titled "Dissatisfaction regarding the food temperature" and notifies the terminal. The user (staff) can then use this feedback to quickly take measures such as reheating the food. In this way, this system helps to improve customer satisfaction.

[0224] The following describes the processing flow.

[0225] Step 1:

[0226] The server acquires video data in real time from cameras installed in the store. The video data is transmitted from multiple cameras installed at each table.

[0227] Step 2:

[0228] The server identifies the visitor's face from the acquired video data using a face region detection algorithm. An expression analysis algorithm is then applied to the identified face region to classify it into a specific emotion category.

[0229] Step 3:

[0230] The terminal transmits audio data collected within the store to a server. This audio data is recorded by microphones installed at each table.

[0231] Step 4:

[0232] The server processes the transmitted audio data through an audio preprocessing engine to remove noise and convert it into a clear signal. Next, it uses an audio processing algorithm to extract audio features and perform emotion estimation based on the audio data.

[0233] Step 5:

[0234] The server integrates emotion categories from images and emotion estimation results from audio, and records the visitor's overall emotional state in a database.

[0235] Step 6:

[0236] The server analyzes accumulated emotional data to identify customer emotional trends and automatically generates reports based on this analysis.

[0237] Step 7:

[0238] The terminal displays the generated reports to store managers and staff. Users (staff) can use this information to implement operational improvement measures in real time.

[0239] (Example 1)

[0240] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0241] In today's service industry, improving customer satisfaction requires understanding customers' emotional states in real time and responding appropriately. However, conventional methods have made it difficult to accurately analyze emotions from customers' facial expressions and voices and effectively provide feedback. This invention solves this problem and provides a means to quickly and accurately detect customer emotions and use that information to improve services.

[0242] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0243] In this invention, the server includes image processing means for detecting a visitor's facial expression, data classification means for classifying the visitor's emotions based on information extracted from the facial expression, and voice processing means for acquiring and processing the visitor's voice. This makes it possible to integrate and analyze emotion data extracted from the customer's facial expression and voice, generate appropriate reports in real time, and improve service.

[0244] "Image processing means" refers to a technology for detecting a visitor's facial expression and extracting facial features from that expression.

[0245] "Data classification means" refers to a method of classifying visitors' emotions into multiple categories based on information obtained through image processing means.

[0246] "Voice processing means" refers to a method of acquiring the voice of a visitor and performing preprocessing on that voice signal.

[0247] "Acoustic analysis means" refers to techniques for estimating a visitor's emotions using the characteristics of their voice.

[0248] "Information analysis means" refers to a process that integrates data obtained from image processing means and audio processing means to comprehensively analyze customer emotions.

[0249] "Information generation means" refers to a function that uses analyzed sentiment data to create report information and provides feedback for service improvement.

[0250] This invention provides a system that analyzes customer emotions in real time in the service industry and improves service quality. This system mainly consists of a server, terminals, and users.

[0251] The server receives video data of visitors from cameras installed in the store. When identifying facial regions from this video data, the server uses an image processing library such as OpenCV. In addition, to analyze facial expressions and classify emotions, it performs emotion classification using machine learning models such as TensorFlow.

[0252] The terminal receives audio data from microphones in the store and sends it to the server. The server analyzes the audio data using an audio processing library such as LibROSA and estimates emotions from the audio features. This analysis is based on the pitch, speed, and volume of the speech.

[0253] Users (e.g., service staff) can view reports generated by the server on their administrator terminals, enabling immediate responses based on the customer's emotional state. This can improve customer satisfaction.

[0254] As a concrete example, consider a situation where a visitor is eating a meal. In this case, the camera captures their facial expression, and the server detects an emotion of "dissatisfaction." Then, the server analyzes the audio captured through the microphone and determines that the visitor said, "The food is cold." As a result, the server generates a report of "dissatisfaction regarding the temperature of the food" and notifies the terminal. Consequently, the user can rationally take action to reheat the food.

[0255] (Example of prompts for a generative AI model)

[0256] "Please describe a system that uses cameras and microphones installed in restaurants to analyze the facial expressions and voices of customers and understand their emotions in real time. Include specific hardware and software names, usage instructions, and examples."

[0257] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0258] Step 1:

[0259] The server receives real-time video data from cameras installed within the store. Using this video data as input, it extracts face regions using the OpenCV library. This process yields output of face coordinates, which determine the position and size of the faces. Specifically, this involves identifying faces frame by frame and preparing the information necessary for subsequent processing.

[0260] Step 2:

[0261] The server analyzes facial expressions using machine learning models such as TensorFlow, based on facial region information. Using the obtained facial coordinates as input data, it classifies expressions into emotional categories such as joy, surprise, and dissatisfaction. The output of this process is a numerical emotion score. Specifically, for example, a score of 0.8 would represent "joy."

[0262] Step 3:

[0263] The device transmits audio data acquired through the microphone to the server. The transmitted audio data is received as input, and the server uses the LibROSA library to extract audio features. The output obtained here consists of numerical acoustic features such as pitch, velocity, and volume. Specifically, if the audio has a higher pitch than normal, that value is measured.

[0264] Step 4:

[0265] The server performs analysis to estimate emotions from speech based on the extracted acoustic features. Using the acoustic features analyzed by LibROSA as input, it uses a machine learning algorithm to estimate emotions. The output of this process is a score for each emotion category. Specifically, emotions such as "excitement" are scored and output based on factors like pitch and speed.

[0266] Step 5:

[0267] The server integrates emotional data from video and audio. It takes the emotional scores from each analysis step as input and combines them to create consistent customer emotional data. This output is a comprehensive emotional profile showing how the customer felt in the store. Specifically, it yields integrated emotional scores such as "Joy 0.8, Excitement 0.6".

[0268] Step 6:

[0269] The server analyzes the integrated emotional profiles and generates reports. It takes the emotional profiles obtained during the integration process as input, analyzes the output, and creates reports that include recommendations for service improvement. Specifically, it aggregates the emotions customers expressed during specific events or with specific products and presents them as areas for improvement in the store.

[0270] (Application Example 1)

[0271] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0272] In today's service industry, improving customer satisfaction requires accurately understanding customer emotions and responding promptly and appropriately. However, traditional systems have struggled to grasp customer intentions and emotions in real time and respond immediately, resulting in limitations in improving customer satisfaction. Furthermore, employees have limitations in their ability to directly read customers' emotions, which hinders effective customer service.

[0273] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0274] In this invention, the server includes an image generation means for detecting the visitor's facial expression, a notification means for notifying an information display device of the emotional data in real time to enable immediate response, and a display means for displaying the analyzed emotional data to facilitate service based on the visitor's state. This allows employees to grasp the customer's emotions in real time and immediately use that information to improve service, thereby increasing customer satisfaction.

[0275] The "image generation means" is a method for detecting a visitor's facial expression using a camera and generating image data in real time.

[0276] An "emotion classification method" is a means of classifying a visitor's emotions into pre-defined categories based on acquired facial expression data.

[0277] "Voice acquisition means" refers to a means for acquiring the voice of a visitor and converting that data into an analyzable format.

[0278] A "voice emotion estimation means" is a voice analysis means used to identify the emotions of a visitor based on acquired voice data.

[0279] "Data analysis means" refers to a method for integrating and comprehensively analyzing facial expression data and voice data.

[0280] A "report generation method" is a means of generating output information by utilizing analyzed sentiment data.

[0281] A "notification means" is a means of transmitting analyzed emotion data to an information display device in real time and immediately notifying the user of related information.

[0282] "Display means" refers to means for visually displaying information so that a user can perform an appropriate service based on emotional data.

[0283] The system that realizes this application example consists of the following elements. The server plays a central role in the customer service support system using smart glasses. First, the server receives video data from a camera installed in the store and uses this data to recognize the expressions of visitors. OpenCV is utilized for image processing, and a face recognition algorithm is applied to identify individual face regions. Based on this face data, emotions are classified using TensorFlow.

[0284] Next, the server receives audio data from a microphone and uses PyDub and Sphinx for audio analysis. In this analysis, features such as the pitch, speed, and volume of the audio are extracted to perform audio emotion estimation. The emotional data thus obtained is integrated by the server and transmitted in real-time to the administrator terminal via Firebase.

[0285] On the other hand, the staff, who is the user, wears smart glasses and can confirm the emotional information of customers in real-time through the display device. This is because the emotional data is displayed on the glasses via the notification means. For example, when a customer places an order with a staff member wearing smart glasses and shows an anxious expression regarding the order content, the information "The customer looks anxious. Order confirmation is required." is displayed on the glasses' display. Based on this information, the user can quickly take actions to eliminate the customer's anxiety.

[0286] Examples of prompt sentences using the generative AI model include "Please explain the method for generating customer emotion data that integrates the customer's expression and voice and can be referred to by employees in real-time." Thus, this system improves the interaction with customers and realizes an improvement in service quality.

[0287] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0288] Step 1:

[0289] The server receives video data in real time from cameras installed in the store. It processes the video data as input using OpenCV, and identifies individual face regions using a face recognition algorithm. This process yields face position information and facial expression data as output.

[0290] Step 2:

[0291] The server uses TensorFlow to classify the facial expression data extracted in the previous step into emotion categories. The input data is facial expression data, and the server analyzes this data to output emotion tags such as "joy" and "dissatisfaction." Based on this data, the server is ready to provide information to the staff.

[0292] Step 3:

[0293] The server receives audio data from microphones in the store. The input audio is preprocessed with PyDub, and speech analysis is performed using Sphinx. Features such as pitch and volume are extracted, and emotion estimation is performed to obtain emotion-related speech features as output.

[0294] Step 4:

[0295] The server integrates facial expression data and voice data. Here, both sets of data are analyzed and the resulting emotional data is structured using Firebase. The output is emotional data formatted for use in decision-making.

[0296] Step 5:

[0297] The server notifies staff members in real time of integrated emotional data via smart glasses. This notification system allows staff to display customer emotional information (the input data) on the smart glasses, enabling immediate responses. The display includes specific customer service advice.

[0298] Step 6:

[0299] Staff members, acting as users, select actions based on customer emotion information displayed on the smart glasses' screen. During interactions with customers, they choose actions to improve service based on the emotion information. This ensures that emotion data, as input, is reflected in practice, leading to improved customer satisfaction.

[0300] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0301] This invention provides a system for accurately and in real time understanding customer emotions in the service industry and improving the quality of service. This system mainly consists of a server, terminals, users, and an emotion engine.

[0302] The server receives video data in real time from cameras installed in the store and applies video processing algorithms to detect visitors' facial expressions. Based on these detected expressions, an emotion classification system categorizes the visitor's emotions. This classification is divided into categories such as joy, surprise, sadness, and dissatisfaction, and the customer's instantaneous emotions are also recorded.

[0303] Simultaneously, the server processes the audio data transmitted from the terminal. The terminal collects audio data through microphones installed in the store and transmits it to the server. The server performs preprocessing such as noise reduction and audio normalization, and estimates the emotion from the audio using an audio emotion estimation method.

[0304] Furthermore, an emotion engine is incorporated into the system, enabling multi-faceted emotion recognition that includes the user's behavioral data and biometric signals. The emotion engine uses a machine learning model to analyze various collected data and accurately recognize the customer's emotional state.

[0305] The server integrates the facial expression data, voice data, and data from the emotion engine and performs analysis. Based on the analysis results, the server generates a report in real-time and transmits it to the terminal. This report can be viewed by administrators and staff and contains the customer's emotion trends and recommended actions for service improvement.

[0306] As a specific example, when a certain visitor expresses dissatisfaction during a meal, the camera captures the expression, and the server classifies it as an "dissatisfaction" emotion. At the same time, if the word "slow" is extracted from the voice data, the emotion engine that processes this information recognizes that the user is dissatisfied with the slowness of the service. This information is immediately generated as a report and notified to the staff's terminal. The user (staff) can quickly implement countermeasures based on this and improve the customer's satisfaction. In this way, this system enables real-time emotion recognition and on-site service improvement, making the experience with customers better.

[0307] The processing flow will be described below.

[0308] Step 1:

[0309] The server receives video data in real-time from cameras installed in the store. The cameras are placed at positions that capture each table where visitors sit.

[0310] Step 2:

[0311] The server applies a face region detection algorithm to the received video data to identify the faces of individual visitors. For the identified face regions, the server analyzes the expressions and classifies the emotions using a deep learning model.

[0312] Step 3:

[0313] The terminal collects the audio of visitors' conversations through microphones in the store and simultaneously transmits that audio data to a server.

[0314] Step 4:

[0315] The server performs noise reduction and normalization on the received audio data. Next, it applies an audio emotion estimation algorithm to estimate the emotion based on the audio.

[0316] Step 5:

[0317] The server integrates emotional data obtained from facial expressions and emotional data obtained from voice. It then activates an emotion engine and uses a machine learning-based model to accurately recognize the user's emotions from the integrated data.

[0318] Step 6:

[0319] The server analyzes the recognized emotion data to understand customer emotion trends and generates a report based on the analysis results.

[0320] Step 7:

[0321] The terminal receives reports sent from the server and displays them to administrators and staff. Based on these reports, users (staff) take appropriate actions in real time to improve customer satisfaction.

[0322] (Example 2)

[0323] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0324] In the service industry, accurately understanding customer emotions in real time and improving service quality is essential. However, conventional systems have been insufficient in analyzing data obtained from video and audio, making it difficult to understand customer emotions from multiple perspectives. Furthermore, it has been difficult to appropriately utilize the analysis results to improve customer service.

[0325] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0326] In this invention, the server includes video processing means for detecting a visitor's facial expression, emotion classification means for classifying the visitor's emotions based on the facial expression, and acoustic data collection means for acquiring the visitor's voice. This enables multifaceted emotional recognition of the customer.

[0327] "Image processing means" refers to technology for detecting visitors' facial expressions in real time and analyzing the image data.

[0328] "Emotion classification means" refers to a technology that classifies and categorizes a visitor's emotions based on the characteristics of their facial expressions.

[0329] "Acoustic data collection means" refers to technology for collecting sounds generated within a store and pre-processing them into a format suitable for analysis.

[0330] "Voice emotion estimation means" refers to a technology for estimating a visitor's emotions from collected voice data.

[0331] "Emotion recognition means" refers to a technology that analyzes multiple data points, including user behavior data and biosignals, from various angles to recognize visitors' emotions with high accuracy.

[0332] "Data analysis means" refers to technology for integrating the results of facial expression, voice, and emotion recognition and performing detailed analysis.

[0333] "Information generation means" refers to technology for generating output information, including measures to improve customer service, based on analyzed emotional data.

[0334] This invention is configured as a system for understanding customer emotions in the service industry in real time and improving the quality of service. This system consists of multiple elements, including a server, terminals, users, and an emotion engine.

[0335] The server receives real-time video data from high-resolution cameras installed within the store. The server uses video processing algorithms to detect visitors' facial expressions and analyzes this data using emotion classification tools. This allows the server to classify visitors' emotions into categories such as "joy," "surprise," "sadness," and "dissatisfaction."

[0336] The terminal collects audio data through multiple microphones installed within the store and transmits it to a server. The server performs noise reduction and speech normalization on the audio data and estimates the customer's emotions using speech emotion estimation tools. These processes can utilize common open-source speech processing libraries or proprietary speech analysis tools.

[0337] Furthermore, the system incorporates an emotion engine that collects and analyzes multifaceted data, including behavioral data and biosignals. The emotion engine utilizes a generative AI model and machine learning to accurately recognize customer emotions. This allows the server to analyze multiple integrated emotional data sets and generate reports based on the results.

[0338] Users (staff) receive reports notified to their terminals and use the analysis results to improve customer service. For example, if a visitor expresses dissatisfaction in the store, the server analyzes their facial expressions and voice data and creates a report so that staff can respond immediately.

[0339] The following is an example of a specific example and prompt: "If a customer in a restaurant expresses dissatisfaction during their meal, explain how you would analyze their emotions in real time and communicate that information to staff to address the issue."

[0340] The advantage of this system lies in its ability to enable real-time emotion recognition, allowing for quick and effective implementation of countermeasures to improve the customer experience.

[0341] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0342] Step 1:

[0343] The terminal uses cameras installed in the store to acquire real-time video data of visitors. This video data becomes the input to the system. The terminal sends this data directly to the server to prepare for subsequent processing. The actions performed here include camera adjustment and motion detection.

[0344] Step 2:

[0345] The server applies a face detection algorithm to the received video data. This identifies face regions from each frame. The output of this process is the coordinate data of each face region. By identifying the position of the faces, the data necessary for subsequent facial expression analysis is prepared.

[0346] Step 3:

[0347] The server applies an expression recognition algorithm to the data after face detection is complete, classifying emotions. In this process, emotions such as joy, surprise, sadness, and dissatisfaction are determined from facial feature points. The output is a label for the emotion category.

[0348] Step 4:

[0349] The terminal collects customer voice data using microphones installed in the store. This voice data is input to a server, where it undergoes pre-processing such as noise reduction. Since the collected data includes conversation and background noise, it is adjusted to extract important voice components.

[0350] Step 5:

[0351] The server receives denoised audio data and applies an audio emotion estimation algorithm to it. Based on the features extracted from the audio, it estimates the customer's emotions. The output is an emotion label from the audio, ready to be integrated with facial expression data.

[0352] Step 6:

[0353] The server integrates emotional information obtained from both facial expression and voice data into an emotion engine. This emotion engine uses a generative AI model and machine learning to perform comprehensive emotional analysis. The analysis results of multiple emotional data are output and transformed into a form that is meaningful to the user.

[0354] Step 7:

[0355] The server generates a report based on the analyzed sentiment data. The output includes customer sentiment trends and suggestions for service improvement. This report is sent to terminals, allowing users (staff) to understand it in real time and use it to improve their responses.

[0356] (Application Example 2)

[0357] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0358] In traditional service industries, it is difficult to instantly grasp and respond to customer emotions, limiting the potential for improving service quality. There is a need for technology that can analyze customer emotions in real time and enable service providers to respond appropriately on the spot.

[0359] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0360] In this invention, the server includes a video generation means for detecting a visitor's facial expression, an emotion classification means for classifying the visitor's emotions based on the facial expression, and an audio acquisition means for acquiring the visitor's sound. This allows service staff to recognize customers' emotions in real time and immediately improve their service.

[0361] The "image generation means" is a mechanism that acquires and processes the video data necessary to detect the facial expressions of visitors.

[0362] An "emotion classification system" is a mechanism that classifies a visitor's emotions based on acquired facial expression data.

[0363] "Sound acquisition means" refers to a mechanism for acquiring sound data of visitors.

[0364] An "acoustic emotion estimation means" is a mechanism that estimates emotions based on acoustic data.

[0365] "Information analysis means" refers to a mechanism that integrates and analyzes emotional data based on facial expressions and sounds.

[0366] An "information generation means" is a mechanism that generates output information using analyzed emotion data.

[0367] A "wearable display device" is a device equipped with a display device for presenting output information to customers.

[0368] This invention is a system that grasps customer emotions in real time in physical stores and improves the quality of service. The server processes video data acquired from cameras installed in the store using video generation means for detecting visitors' facial expressions and performs facial recognition. Furthermore, it collects acoustic data from microphones installed in the store using acoustic acquisition means for acquiring acoustic data of visitors and estimates acoustic emotions.

[0369] The server uses information analysis tools to comprehensively analyze this data and recognize the customer's emotional state with high accuracy. From the analyzed emotional data, an information generation tool generates output information, which is then presented to the service provider on the spot via a wearable display device. This information provides crucial clues for the service provider to take quick action.

[0370] Specific hardware includes wearable display devices such as smart glasses. Software used includes OpenCV for video processing, librosa for audio processing, and scikit-learn for machine learning models. This enables real-time customer emotion recognition and rapid service improvement. A concrete application of the invention is that, for example, if a customer shows dissatisfaction while waiting, their facial expression and sound can be used to recognize their emotion as "dissatisfaction," allowing staff to respond quickly and improve customer satisfaction.

[0371] An example of a specific prompt for the generating AI model is, "Based on the visitor's facial expressions and audio data, please list the most suitable recommended actions to improve customer service on the spot." This allows service staff to quickly implement appropriate countermeasures.

[0372] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0373] Step 1:

[0374] The server acquires video data from cameras installed within the store. The input is real-time video data, and the output is image data including the facial regions of visitors. The video generation method uses OpenCV to perform face recognition and extracts facial regions of particular interest.

[0375] Step 2:

[0376] The server acquires acoustic data from microphones installed within the store. The input is acoustic data including ambient noise, and the output is pre-processed acoustic data. The acoustic acquisition method uses librosa to remove noise and normalize the acoustic signal.

[0377] Step 3:

[0378] The server classifies emotions based on facial expression data. The input is image data of the face region, and the output is the classified emotion category (e.g., joy, surprise, dissatisfaction). The emotion classification means analyzes facial landmarks and estimates emotions using a machine learning model.

[0379] Step 4:

[0380] The server estimates acoustic emotion based on acoustic data. The input is normalized acoustic data, and the output is the estimated emotion category. The acoustic emotion estimation means performs emotion estimation based on voice tone and phoneme information.

[0381] Step 5:

[0382] The server integrates and analyzes emotional information obtained from facial expression data and acoustic data. The input is multiple emotional data, and the output is the overall emotional state of the customer. The information analysis means integrates different emotional indicators with weights to generate an overall picture.

[0383] Step 6:

[0384] The server generates output information based on the analysis results. The input is the overall emotional state, and the output is a report containing recommended actions for improving customer service. The information generation means uses a generation AI model to create a list based on prompt sentences.

[0385] Step 7:

[0386] The user receives output information generated through a wearable display device. The input is a real-time report, and the output manifests as a rapid action taken by the service provider. Display devices such as smart glasses visually notify the service provider of the emotional state.

[0387] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0388] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0389] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0390] [Third Embodiment]

[0391] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0392] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0393] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0394] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0395] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0396] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0397] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0398] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0399] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0400] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0401] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0402] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0403] This invention provides a system for service industries such as restaurants that can grasp customer emotions in real time and improve the quality of service. This system mainly consists of a server, terminals, and users.

[0404] The server receives video data from cameras to detect visitors' facial expressions. Cameras are installed in multiple locations within the store and can capture visitors' facial expressions. The server recognizes facial regions from the video data and applies an expression analysis algorithm to classify the visitor's emotions. This classification can be divided into categories such as joy, surprise, and dissatisfaction.

[0405] Furthermore, the server receives audio data from microphones installed within the store. The terminals are responsible for transmitting the audio data to the server. The server analyzes the audio data, extracts specific voice features, and estimates the emotion. This analysis is based on factors such as the pitch, speed, and volume of the voice.

[0406] The server integrates emotional data from video and audio and records it in a database. This ensures that customer emotional data is stored in a time-synchronized manner. Based on this data, the server performs data analysis and generates reports.

[0407] The generated report is sent to the administrator's terminal, where store managers and staff can review it. This report includes customer sentiment trends, frequent patterns, and recommendations for service improvement.

[0408] As a concrete example, suppose a visitor is eating a meal, and the camera captures their facial expression, and the server detects an emotion of "dissatisfaction." At the same time, the server analyzes the audio picked up by the microphone and detects a statement such as "the food is cold." Based on this information, the server generates a report titled "Dissatisfaction regarding the food temperature" and notifies the terminal. The user (staff) can then use this feedback to quickly take measures such as reheating the food. In this way, this system helps to improve customer satisfaction.

[0409] The following describes the processing flow.

[0410] Step 1:

[0411] The server acquires video data in real time from cameras installed in the store. The video data is transmitted from multiple cameras installed at each table.

[0412] Step 2:

[0413] The server identifies the visitor's face from the acquired video data using a face region detection algorithm. An expression analysis algorithm is then applied to the identified face region to classify it into a specific emotion category.

[0414] Step 3:

[0415] The terminal transmits audio data collected within the store to a server. This audio data is recorded by microphones installed at each table.

[0416] Step 4:

[0417] The server processes the transmitted audio data through an audio preprocessing engine to remove noise and convert it into a clear signal. Next, it uses an audio processing algorithm to extract audio features and perform emotion estimation based on the audio data.

[0418] Step 5:

[0419] The server integrates emotion categories from images and emotion estimation results from audio, and records the visitor's overall emotional state in a database.

[0420] Step 6:

[0421] The server analyzes accumulated emotional data to identify customer emotional trends and automatically generates reports based on this analysis.

[0422] Step 7:

[0423] The terminal displays the generated reports to store managers and staff. Users (staff) can use this information to implement operational improvement measures in real time.

[0424] (Example 1)

[0425] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0426] In today's service industry, improving customer satisfaction requires understanding customers' emotional states in real time and responding appropriately. However, conventional methods have made it difficult to accurately analyze emotions from customers' facial expressions and voices and effectively provide feedback. This invention solves this problem and provides a means to quickly and accurately detect customer emotions and use that information to improve services.

[0427] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0428] In this invention, the server includes image processing means for detecting a visitor's facial expression, data classification means for classifying the visitor's emotions based on information extracted from the facial expression, and voice processing means for acquiring and processing the visitor's voice. This makes it possible to integrate and analyze emotion data extracted from the customer's facial expression and voice, generate appropriate reports in real time, and improve service.

[0429] "Image processing means" refers to a technology for detecting a visitor's facial expression and extracting facial features from that expression.

[0430] "Data classification means" refers to a method of classifying visitors' emotions into multiple categories based on information obtained through image processing means.

[0431] "Voice processing means" refers to a method of acquiring the voice of a visitor and performing preprocessing on that voice signal.

[0432] "Acoustic analysis means" refers to techniques for estimating a visitor's emotions using the characteristics of their voice.

[0433] "Information analysis means" refers to a process that integrates data obtained from image processing means and audio processing means to comprehensively analyze customer emotions.

[0434] "Information generation means" refers to a function that uses analyzed sentiment data to create report information and provides feedback for service improvement.

[0435] This invention provides a system that analyzes customer emotions in real time in the service industry and improves service quality. This system mainly consists of a server, terminals, and users.

[0436] The server receives video data of visitors from cameras installed in the store. When identifying facial regions from this video data, the server uses an image processing library such as OpenCV. In addition, to analyze facial expressions and classify emotions, it performs emotion classification using machine learning models such as TensorFlow.

[0437] The terminal receives audio data from microphones in the store and sends it to the server. The server analyzes the audio data using an audio processing library such as LibROSA and estimates emotions from the audio features. This analysis is based on the pitch, speed, and volume of the speech.

[0438] Users (e.g., service staff) can view reports generated by the server on their administrator terminals, enabling immediate responses based on the customer's emotional state. This can improve customer satisfaction.

[0439] As a concrete example, consider a situation where a visitor is eating a meal. In this case, the camera captures their facial expression, and the server detects an emotion of "dissatisfaction." Then, the server analyzes the audio captured through the microphone and determines that the visitor said, "The food is cold." As a result, the server generates a report of "dissatisfaction regarding the temperature of the food" and notifies the terminal. Consequently, the user can rationally take action to reheat the food.

[0440] (Example of prompts for a generative AI model)

[0441] "Please describe a system that uses cameras and microphones installed in restaurants to analyze the facial expressions and voices of customers and understand their emotions in real time. Include specific hardware and software names, usage instructions, and examples."

[0442] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0443] Step 1:

[0444] The server receives real-time video data from cameras installed within the store. Using this video data as input, it extracts face regions using the OpenCV library. This process yields output of face coordinates, which determine the position and size of the faces. Specifically, this involves identifying faces frame by frame and preparing the information necessary for subsequent processing.

[0445] Step 2:

[0446] The server analyzes facial expressions using machine learning models such as TensorFlow, based on facial region information. Using the obtained facial coordinates as input data, it classifies expressions into emotional categories such as joy, surprise, and dissatisfaction. The output of this process is a numerical emotion score. Specifically, for example, a score of 0.8 would represent "joy."

[0447] Step 3:

[0448] The device transmits audio data acquired through the microphone to the server. The transmitted audio data is received as input, and the server uses the LibROSA library to extract audio features. The output obtained here consists of numerical acoustic features such as pitch, velocity, and volume. Specifically, if the audio has a higher pitch than normal, that value is measured.

[0449] Step 4:

[0450] The server performs analysis to estimate emotions from speech based on the extracted acoustic features. Using the acoustic features analyzed by LibROSA as input, it uses a machine learning algorithm to estimate emotions. The output of this process is a score for each emotion category. Specifically, emotions such as "excitement" are scored and output based on factors like pitch and speed.

[0451] Step 5:

[0452] The server integrates emotional data from video and audio. It takes the emotional scores from each analysis step as input and combines them to create consistent customer emotional data. This output is a comprehensive emotional profile showing how the customer felt in the store. Specifically, it yields integrated emotional scores such as "Joy 0.8, Excitement 0.6".

[0453] Step 6:

[0454] The server analyzes the integrated emotional profiles and generates reports. It takes the emotional profiles obtained during the integration process as input, analyzes the output, and creates reports that include recommendations for service improvement. Specifically, it aggregates the emotions customers expressed during specific events or with specific products and presents them as areas for improvement in the store.

[0455] (Application Example 1)

[0456] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0457] In today's service industry, improving customer satisfaction requires accurately understanding customer emotions and responding promptly and appropriately. However, traditional systems have struggled to grasp customer intentions and emotions in real time and respond immediately, resulting in limitations in improving customer satisfaction. Furthermore, employees have limitations in their ability to directly read customers' emotions, which hinders effective customer service.

[0458] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0459] In this invention, the server includes an image generation means for detecting the visitor's facial expression, a notification means for notifying an information display device of the emotional data in real time to enable immediate response, and a display means for displaying the analyzed emotional data to facilitate service based on the visitor's state. This allows employees to grasp the customer's emotions in real time and immediately use that information to improve service, thereby increasing customer satisfaction.

[0460] The "image generation means" is a method for detecting a visitor's facial expression using a camera and generating image data in real time.

[0461] An "emotion classification method" is a means of classifying a visitor's emotions into pre-defined categories based on acquired facial expression data.

[0462] "Voice acquisition means" refers to a means for acquiring the voice of a visitor and converting that data into an analyzable format.

[0463] A "voice emotion estimation means" is a voice analysis means used to identify the emotions of a visitor based on acquired voice data.

[0464] "Data analysis means" refers to a method for integrating and comprehensively analyzing facial expression data and voice data.

[0465] A "report generation method" is a means of generating output information by utilizing analyzed sentiment data.

[0466] A "notification means" is a means of transmitting analyzed emotion data to an information display device in real time and immediately notifying the user of related information.

[0467] A "display means" is a means of visually displaying information so that users can perform appropriate services based on emotional data.

[0468] The system that realizes this application consists of the following elements: The server plays a central role in the customer service support system using smart glasses. First, the server receives video data from cameras installed in the store and uses this data to recognize the facial expressions of visitors. OpenCV is used for image processing, and a face recognition algorithm is applied to identify individual face regions. Based on this face data, emotions are classified using TensorFlow.

[0469] Next, the server receives audio data from the microphone and uses PyDub and Sphinx for speech analysis. This analysis extracts features such as pitch, speed, and volume from the speech and performs speech emotion estimation. The resulting emotion data is then integrated by the server and sent in real time to the administrator terminal via Firebase.

[0470] On the other hand, the staff, who are the users, wear smart glasses and can check customer emotional information in real time through the display device. This is because emotional data is displayed on the glasses via a notification system. For example, when a customer places an order with a staff member wearing smart glasses, if the customer shows an anxious expression regarding the order, this information will be displayed on the glasses' screen as, "The customer appears anxious. The order needs to be confirmed." Based on this information, the user can quickly take action to alleviate the customer's anxiety.

[0471] An example of a prompt using the generative AI model might be, "Please explain how to generate customer sentiment data that integrates customer facial expressions and voice, and can be referenced by employees in real time." In this way, the system improves customer interaction and enhances the quality of service.

[0472] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0473] Step 1:

[0474] The server receives video data in real time from cameras installed in the store. It processes the video data as input using OpenCV, and identifies individual face regions using a face recognition algorithm. This process yields face position information and facial expression data as output.

[0475] Step 2:

[0476] The server uses TensorFlow to classify the facial expression data extracted in the previous step into emotion categories. The input data is facial expression data, and the server analyzes this data to output emotion tags such as "joy" and "dissatisfaction." Based on this data, the server is ready to provide information to the staff.

[0477] Step 3:

[0478] The server receives audio data from microphones in the store. The input audio is preprocessed with PyDub, and speech analysis is performed using Sphinx. Features such as pitch and volume are extracted, and emotion estimation is performed to obtain emotion-related speech features as output.

[0479] Step 4:

[0480] The server integrates facial expression data and voice data. Here, both sets of data are analyzed and the resulting emotional data is structured using Firebase. The output is emotional data formatted for use in decision-making.

[0481] Step 5:

[0482] The server notifies staff members in real time of integrated emotional data via smart glasses. This notification system allows staff to display customer emotional information (the input data) on the smart glasses, enabling immediate responses. The display includes specific customer service advice.

[0483] Step 6:

[0484] Staff members, acting as users, select actions based on customer emotion information displayed on the smart glasses' screen. During interactions with customers, they choose actions to improve service based on the emotion information. This ensures that emotion data, as input, is reflected in practice, leading to improved customer satisfaction.

[0485] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0486] This invention provides a system for accurately and in real time understanding customer emotions in the service industry and improving the quality of service. This system mainly consists of a server, terminals, users, and an emotion engine.

[0487] The server receives video data in real time from cameras installed in the store and applies video processing algorithms to detect visitors' facial expressions. Based on these detected expressions, an emotion classification system categorizes the visitor's emotions. This classification is divided into categories such as joy, surprise, sadness, and dissatisfaction, and the customer's instantaneous emotions are also recorded.

[0488] Simultaneously, the server processes the audio data transmitted from the terminal. The terminal collects audio data through microphones installed in the store and transmits it to the server. The server performs preprocessing such as noise reduction and audio normalization, and estimates the emotion from the audio using an audio emotion estimation method.

[0489] Furthermore, an emotion engine is integrated into the system, enabling multifaceted emotion recognition that includes user behavior data and biometric signals. The emotion engine uses machine learning models to analyze various collected data and recognize the customer's emotional state with high accuracy.

[0490] The server integrates and analyzes facial expression data, voice data, and data from the emotion engine. Based on this analysis, the server generates a report in real time and sends it to the terminal. This report is viewable by administrators and staff and contains customer emotion trends and recommended actions for service improvement.

[0491] For example, if a visitor expresses dissatisfaction during a meal, the camera captures their facial expression, and the server classifies it as "dissatisfaction." Simultaneously, if the word "slow" is extracted from the audio data, the emotion engine, processing this information, recognizes that the user is dissatisfied with the slow service. This information is immediately generated as a report and notified to the staff's terminal. Based on this, the user (staff) can quickly implement countermeasures to improve customer satisfaction. In this way, this system enables real-time emotion recognition and on-the-spot service improvement, resulting in a better customer experience.

[0492] The following describes the processing flow.

[0493] Step 1:

[0494] The server receives video data in real time from cameras installed within the store. The cameras are positioned to capture each table where customers are seated.

[0495] Step 2:

[0496] The server applies a face region detection algorithm to the received video data to identify the faces of individual visitors. A deep learning model is then used to analyze the facial expressions of the identified face regions and classify the emotions.

[0497] Step 3:

[0498] The terminal collects the audio of visitors' conversations through microphones in the store and simultaneously transmits that audio data to a server.

[0499] Step 4:

[0500] The server performs noise reduction and normalization on the received audio data. Next, it applies an audio emotion estimation algorithm to estimate the emotion based on the audio.

[0501] Step 5:

[0502] The server integrates emotional data obtained from facial expressions and emotional data obtained from voice. It then activates an emotion engine and uses a machine learning-based model to accurately recognize the user's emotions from the integrated data.

[0503] Step 6:

[0504] The server analyzes the recognized emotion data to understand customer emotion trends and generates a report based on the analysis results.

[0505] Step 7:

[0506] The terminal receives reports sent from the server and displays them to administrators and staff. Based on these reports, users (staff) take appropriate actions in real time to improve customer satisfaction.

[0507] (Example 2)

[0508] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0509] In the service industry, accurately understanding customer emotions in real time and improving service quality is essential. However, conventional systems have been insufficient in analyzing data obtained from video and audio, making it difficult to understand customer emotions from multiple perspectives. Furthermore, it has been difficult to appropriately utilize the analysis results to improve customer service.

[0510] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0511] In this invention, the server includes video processing means for detecting a visitor's facial expression, emotion classification means for classifying the visitor's emotions based on the facial expression, and acoustic data collection means for acquiring the visitor's voice. This enables multifaceted emotional recognition of the customer.

[0512] "Image processing means" refers to technology for detecting visitors' facial expressions in real time and analyzing the image data.

[0513] "Emotion classification means" refers to a technology that classifies and categorizes a visitor's emotions based on the characteristics of their facial expressions.

[0514] "Acoustic data collection means" refers to technology for collecting sounds generated within a store and pre-processing them into a format suitable for analysis.

[0515] "Voice emotion estimation means" refers to a technology for estimating a visitor's emotions from collected voice data.

[0516] "Emotion recognition means" refers to a technology that analyzes multiple data points, including user behavior data and biosignals, from various angles to recognize visitors' emotions with high accuracy.

[0517] "Data analysis means" refers to technology for integrating the results of facial expression, voice, and emotion recognition and performing detailed analysis.

[0518] "Information generation means" refers to technology for generating output information, including measures to improve customer service, based on analyzed emotional data.

[0519] This invention is configured as a system for understanding customer emotions in the service industry in real time and improving the quality of service. This system consists of multiple elements, including a server, terminals, users, and an emotion engine.

[0520] The server receives real-time video data from high-resolution cameras installed within the store. The server uses video processing algorithms to detect visitors' facial expressions and analyzes this data using emotion classification tools. This allows the server to classify visitors' emotions into categories such as "joy," "surprise," "sadness," and "dissatisfaction."

[0521] The terminal collects audio data through multiple microphones installed within the store and transmits it to a server. The server performs noise reduction and speech normalization on the audio data and estimates the customer's emotions using speech emotion estimation tools. These processes can utilize common open-source speech processing libraries or proprietary speech analysis tools.

[0522] Furthermore, the system incorporates an emotion engine that collects and analyzes multifaceted data, including behavioral data and biosignals. The emotion engine utilizes a generative AI model and machine learning to accurately recognize customer emotions. This allows the server to analyze multiple integrated emotional data sets and generate reports based on the results.

[0523] Users (staff) receive reports notified to their terminals and use the analysis results to improve customer service. For example, if a visitor expresses dissatisfaction in the store, the server analyzes their facial expressions and voice data and creates a report so that staff can respond immediately.

[0524] The following is an example of a specific example and prompt: "If a customer in a restaurant expresses dissatisfaction during their meal, explain how you would analyze their emotions in real time and communicate that information to staff to address the issue."

[0525] The advantage of this system lies in its ability to enable real-time emotion recognition, allowing for quick and effective implementation of countermeasures to improve the customer experience.

[0526] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0527] Step 1:

[0528] The terminal uses cameras installed in the store to acquire real-time video data of visitors. This video data becomes the input to the system. The terminal sends this data directly to the server to prepare for subsequent processing. The actions performed here include camera adjustment and motion detection.

[0529] Step 2:

[0530] The server applies a face detection algorithm to the received video data. This identifies face regions from each frame. The output of this process is the coordinate data of each face region. By identifying the position of the faces, the data necessary for subsequent facial expression analysis is prepared.

[0531] Step 3:

[0532] The server applies an expression recognition algorithm to the data after face detection is complete, classifying emotions. In this process, emotions such as joy, surprise, sadness, and dissatisfaction are determined from facial feature points. The output is a label for the emotion category.

[0533] Step 4:

[0534] The terminal collects customer voice data using microphones installed in the store. This voice data is input to a server, where it undergoes pre-processing such as noise reduction. Since the collected data includes conversation and background noise, it is adjusted to extract important voice components.

[0535] Step 5:

[0536] The server receives denoised audio data and applies an audio emotion estimation algorithm to it. Based on the features extracted from the audio, it estimates the customer's emotions. The output is an emotion label from the audio, ready to be integrated with facial expression data.

[0537] Step 6:

[0538] The server integrates emotional information obtained from both facial expression and voice data into an emotion engine. This emotion engine uses a generative AI model and machine learning to perform comprehensive emotional analysis. The analysis results of multiple emotional data are output and transformed into a form that is meaningful to the user.

[0539] Step 7:

[0540] The server generates a report based on the analyzed sentiment data. The output includes customer sentiment trends and suggestions for service improvement. This report is sent to terminals, allowing users (staff) to understand it in real time and use it to improve their responses.

[0541] (Application Example 2)

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

[0543] In traditional service industries, it is difficult to instantly grasp and respond to customer emotions, limiting the potential for improving service quality. There is a need for technology that can analyze customer emotions in real time and enable service providers to respond appropriately on the spot.

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

[0545] In this invention, the server includes a video generation means for detecting a visitor's facial expression, an emotion classification means for classifying the visitor's emotions based on the facial expression, and an audio acquisition means for acquiring the visitor's sound. This allows service staff to recognize customers' emotions in real time and immediately improve their service.

[0546] The "image generation means" is a mechanism that acquires and processes the video data necessary to detect the facial expressions of visitors.

[0547] An "emotion classification system" is a mechanism that classifies a visitor's emotions based on acquired facial expression data.

[0548] "Sound acquisition means" refers to a mechanism for acquiring sound data of visitors.

[0549] An "acoustic emotion estimation means" is a mechanism that estimates emotions based on acoustic data.

[0550] "Information analysis means" refers to a mechanism that integrates and analyzes emotional data based on facial expressions and sounds.

[0551] An "information generation means" is a mechanism that generates output information using analyzed emotion data.

[0552] A "wearable display device" is a device equipped with a display device for presenting output information to customers.

[0553] This invention is a system that grasps customer emotions in real time in physical stores and improves the quality of service. The server processes video data acquired from cameras installed in the store using video generation means for detecting visitors' facial expressions and performs facial recognition. Furthermore, it collects acoustic data from microphones installed in the store using acoustic acquisition means for acquiring acoustic data of visitors and estimates acoustic emotions.

[0554] The server uses information analysis tools to comprehensively analyze this data and recognize the customer's emotional state with high accuracy. From the analyzed emotional data, an information generation tool generates output information, which is then presented to the service provider on the spot via a wearable display device. This information provides crucial clues for the service provider to take quick action.

[0555] Specific hardware includes wearable display devices such as smart glasses. Software used includes OpenCV for video processing, librosa for audio processing, and scikit-learn for machine learning models. This enables real-time customer emotion recognition and rapid service improvement. A concrete application of the invention is that, for example, if a customer shows dissatisfaction while waiting, their facial expression and sound can be used to recognize their emotion as "dissatisfaction," allowing staff to respond quickly and improve customer satisfaction.

[0556] An example of a specific prompt for the generating AI model is, "Based on the visitor's facial expressions and audio data, please list the most suitable recommended actions to improve customer service on the spot." This allows service staff to quickly implement appropriate countermeasures.

[0557] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0558] Step 1:

[0559] The server acquires video data from cameras installed within the store. The input is real-time video data, and the output is image data including the facial regions of visitors. The video generation method uses OpenCV to perform face recognition and extracts facial regions of particular interest.

[0560] Step 2:

[0561] The server acquires acoustic data from microphones installed within the store. The input is acoustic data including ambient noise, and the output is pre-processed acoustic data. The acoustic acquisition method uses librosa to remove noise and normalize the acoustic signal.

[0562] Step 3:

[0563] The server classifies emotions based on facial expression data. The input is image data of the face region, and the output is the classified emotion category (e.g., joy, surprise, dissatisfaction). The emotion classification means analyzes facial landmarks and estimates emotions using a machine learning model.

[0564] Step 4:

[0565] The server estimates acoustic emotion based on acoustic data. The input is normalized acoustic data, and the output is the estimated emotion category. The acoustic emotion estimation means performs emotion estimation based on voice tone and phoneme information.

[0566] Step 5:

[0567] The server integrates and analyzes emotional information obtained from facial expression data and acoustic data. The input is multiple emotional data, and the output is the overall emotional state of the customer. The information analysis means integrates different emotional indicators with weights to generate an overall picture.

[0568] Step 6:

[0569] The server generates output information based on the analysis results. The input is the overall emotional state, and the output is a report containing recommended actions for improving customer service. The information generation means uses a generation AI model to create a list based on prompt sentences.

[0570] Step 7:

[0571] The user receives output information generated through a wearable display device. The input is a real-time report, and the output manifests as a rapid action taken by the service provider. Display devices such as smart glasses visually notify the service provider of the emotional state.

[0572] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0573] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0574] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0575] [Fourth Embodiment]

[0576] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0577] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0578] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0579] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0580] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0581] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0582] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0583] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0584] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0585] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0586] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0587] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0588] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0589] This invention provides a system for service industries such as restaurants that can grasp customer emotions in real time and improve the quality of service. This system mainly consists of a server, terminals, and users.

[0590] The server receives video data from cameras to detect visitors' facial expressions. Cameras are installed in multiple locations within the store and can capture visitors' facial expressions. The server recognizes facial regions from the video data and applies an expression analysis algorithm to classify the visitor's emotions. This classification can be divided into categories such as joy, surprise, and dissatisfaction.

[0591] Furthermore, the server receives audio data from microphones installed within the store. The terminals are responsible for transmitting the audio data to the server. The server analyzes the audio data, extracts specific voice features, and estimates the emotion. This analysis is based on factors such as the pitch, speed, and volume of the voice.

[0592] The server integrates emotional data from video and audio and records it in a database. This ensures that customer emotional data is stored in a time-synchronized manner. Based on this data, the server performs data analysis and generates reports.

[0593] The generated report is sent to the administrator's terminal, where store managers and staff can review it. This report includes customer sentiment trends, frequent patterns, and recommendations for service improvement.

[0594] As a concrete example, suppose a visitor is eating a meal, and the camera captures their facial expression, and the server detects an emotion of "dissatisfaction." At the same time, the server analyzes the audio picked up by the microphone and detects a statement such as "the food is cold." Based on this information, the server generates a report titled "Dissatisfaction regarding the food temperature" and notifies the terminal. The user (staff) can then use this feedback to quickly take measures such as reheating the food. In this way, this system helps to improve customer satisfaction.

[0595] The following describes the processing flow.

[0596] Step 1:

[0597] The server acquires video data in real time from cameras installed in the store. The video data is transmitted from multiple cameras installed at each table.

[0598] Step 2:

[0599] The server identifies the visitor's face from the acquired video data using a face region detection algorithm. An expression analysis algorithm is then applied to the identified face region to classify it into a specific emotion category.

[0600] Step 3:

[0601] The terminal transmits audio data collected within the store to a server. This audio data is recorded by microphones installed at each table.

[0602] Step 4:

[0603] The server processes the transmitted audio data through an audio preprocessing engine to remove noise and convert it into a clear signal. Next, it uses an audio processing algorithm to extract audio features and perform emotion estimation based on the audio data.

[0604] Step 5:

[0605] The server integrates emotion categories from images and emotion estimation results from audio, and records the visitor's overall emotional state in a database.

[0606] Step 6:

[0607] The server analyzes accumulated emotional data to identify customer emotional trends and automatically generates reports based on this analysis.

[0608] Step 7:

[0609] The terminal displays the generated reports to store managers and staff. Users (staff) can use this information to implement operational improvement measures in real time.

[0610] (Example 1)

[0611] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0612] In today's service industry, improving customer satisfaction requires understanding customers' emotional states in real time and responding appropriately. However, conventional methods have made it difficult to accurately analyze emotions from customers' facial expressions and voices and effectively provide feedback. This invention solves this problem and provides a means to quickly and accurately detect customer emotions and use that information to improve services.

[0613] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0614] In this invention, the server includes image processing means for detecting a visitor's facial expression, data classification means for classifying the visitor's emotions based on information extracted from the facial expression, and voice processing means for acquiring and processing the visitor's voice. This makes it possible to integrate and analyze emotion data extracted from the customer's facial expression and voice, generate appropriate reports in real time, and improve service.

[0615] "Image processing means" refers to a technology for detecting a visitor's facial expression and extracting facial features from that expression.

[0616] "Data classification means" refers to a method of classifying visitors' emotions into multiple categories based on information obtained through image processing means.

[0617] "Voice processing means" refers to a method of acquiring the voice of a visitor and performing preprocessing on that voice signal.

[0618] "Acoustic analysis means" refers to techniques for estimating a visitor's emotions using the characteristics of their voice.

[0619] "Information analysis means" refers to a process that integrates data obtained from image processing means and audio processing means to comprehensively analyze customer emotions.

[0620] "Information generation means" refers to a function that uses analyzed sentiment data to create report information and provides feedback for service improvement.

[0621] This invention provides a system that analyzes customer emotions in real time in the service industry and improves service quality. This system mainly consists of a server, terminals, and users.

[0622] The server receives video data of visitors from cameras installed in the store. When identifying facial regions from this video data, the server uses an image processing library such as OpenCV. In addition, to analyze facial expressions and classify emotions, it performs emotion classification using machine learning models such as TensorFlow.

[0623] The terminal receives audio data from microphones in the store and sends it to the server. The server analyzes the audio data using an audio processing library such as LibROSA and estimates emotions from the audio features. This analysis is based on the pitch, speed, and volume of the speech.

[0624] Users (e.g., service staff) can view reports generated by the server on their administrator terminals, enabling immediate responses based on the customer's emotional state. This can improve customer satisfaction.

[0625] As a concrete example, consider a situation where a visitor is eating a meal. In this case, the camera captures their facial expression, and the server detects an emotion of "dissatisfaction." Then, the server analyzes the audio captured through the microphone and determines that the visitor said, "The food is cold." As a result, the server generates a report of "dissatisfaction regarding the temperature of the food" and notifies the terminal. Consequently, the user can rationally take action to reheat the food.

[0626] (Example of prompts for a generative AI model)

[0627] "Please describe a system that uses cameras and microphones installed in restaurants to analyze the facial expressions and voices of customers and understand their emotions in real time. Include specific hardware and software names, usage instructions, and examples."

[0628] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0629] Step 1:

[0630] The server receives real-time video data from cameras installed within the store. Using this video data as input, it extracts face regions using the OpenCV library. This process yields output of face coordinates, which determine the position and size of the faces. Specifically, this involves identifying faces frame by frame and preparing the information necessary for subsequent processing.

[0631] Step 2:

[0632] The server analyzes facial expressions using machine learning models such as TensorFlow, based on facial region information. Using the obtained facial coordinates as input data, it classifies expressions into emotional categories such as joy, surprise, and dissatisfaction. The output of this process is a numerical emotion score. Specifically, for example, a score of 0.8 would represent "joy."

[0633] Step 3:

[0634] The device transmits audio data acquired through the microphone to the server. The transmitted audio data is received as input, and the server uses the LibROSA library to extract audio features. The output obtained here consists of numerical acoustic features such as pitch, velocity, and volume. Specifically, if the audio has a higher pitch than normal, that value is measured.

[0635] Step 4:

[0636] The server performs analysis to estimate emotions from speech based on the extracted acoustic features. Using the acoustic features analyzed by LibROSA as input, it uses a machine learning algorithm to estimate emotions. The output of this process is a score for each emotion category. Specifically, emotions such as "excitement" are scored and output based on factors like pitch and speed.

[0637] Step 5:

[0638] The server integrates emotional data from video and audio. It takes the emotional scores from each analysis step as input and combines them to create consistent customer emotional data. This output is a comprehensive emotional profile showing how the customer felt in the store. Specifically, it yields integrated emotional scores such as "Joy 0.8, Excitement 0.6".

[0639] Step 6:

[0640] The server analyzes the integrated emotional profiles and generates reports. It takes the emotional profiles obtained during the integration process as input, analyzes the output, and creates reports that include recommendations for service improvement. Specifically, it aggregates the emotions customers expressed during specific events or with specific products and presents them as areas for improvement in the store.

[0641] (Application Example 1)

[0642] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0643] In today's service industry, improving customer satisfaction requires accurately understanding customer emotions and responding promptly and appropriately. However, traditional systems have struggled to grasp customer intentions and emotions in real time and respond immediately, resulting in limitations in improving customer satisfaction. Furthermore, employees have limitations in their ability to directly read customers' emotions, which hinders effective customer service.

[0644] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0645] In this invention, the server includes an image generation means for detecting the visitor's facial expression, a notification means for notifying an information display device of the emotional data in real time to enable immediate response, and a display means for displaying the analyzed emotional data to facilitate service based on the visitor's state. This allows employees to grasp the customer's emotions in real time and immediately use that information to improve service, thereby increasing customer satisfaction.

[0646] The "image generation means" is a method for detecting a visitor's facial expression using a camera and generating image data in real time.

[0647] An "emotion classification method" is a means of classifying a visitor's emotions into pre-defined categories based on acquired facial expression data.

[0648] "Voice acquisition means" refers to a means for acquiring the voice of a visitor and converting that data into an analyzable format.

[0649] A "voice emotion estimation means" is a voice analysis means used to identify the emotions of a visitor based on acquired voice data.

[0650] "Data analysis means" refers to a method for integrating and comprehensively analyzing facial expression data and voice data.

[0651] A "report generation method" is a means of generating output information by utilizing analyzed sentiment data.

[0652] A "notification means" is a means of transmitting analyzed emotion data to an information display device in real time and immediately notifying the user of related information.

[0653] A "display means" is a means of visually displaying information so that users can perform appropriate services based on emotional data.

[0654] The system that realizes this application consists of the following elements: The server plays a central role in the customer service support system using smart glasses. First, the server receives video data from cameras installed in the store and uses this data to recognize the facial expressions of visitors. OpenCV is used for image processing, and a face recognition algorithm is applied to identify individual face regions. Based on this face data, emotions are classified using TensorFlow.

[0655] Next, the server receives audio data from the microphone and uses PyDub and Sphinx for speech analysis. This analysis extracts features such as pitch, speed, and volume from the speech and performs speech emotion estimation. The resulting emotion data is then integrated by the server and sent in real time to the administrator terminal via Firebase.

[0656] On the other hand, the staff, who are the users, wear smart glasses and can check customer emotional information in real time through the display device. This is because emotional data is displayed on the glasses via a notification system. For example, when a customer places an order with a staff member wearing smart glasses, if the customer shows an anxious expression regarding the order, this information will be displayed on the glasses' screen as, "The customer appears anxious. The order needs to be confirmed." Based on this information, the user can quickly take action to alleviate the customer's anxiety.

[0657] An example of a prompt using the generative AI model might be, "Please explain how to generate customer sentiment data that integrates customer facial expressions and voice, and can be referenced by employees in real time." In this way, the system improves customer interaction and enhances the quality of service.

[0658] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0659] Step 1:

[0660] The server receives video data in real time from cameras installed in the store. It processes the video data as input using OpenCV, and identifies individual face regions using a face recognition algorithm. This process yields face position information and facial expression data as output.

[0661] Step 2:

[0662] The server uses TensorFlow to classify the facial expression data extracted in the previous step into emotion categories. The input data is facial expression data, and the server analyzes this data to output emotion tags such as "joy" and "dissatisfaction." Based on this data, the server is ready to provide information to the staff.

[0663] Step 3:

[0664] The server receives audio data from microphones in the store. The input audio is preprocessed with PyDub, and speech analysis is performed using Sphinx. Features such as pitch and volume are extracted, and emotion estimation is performed to obtain emotion-related speech features as output.

[0665] Step 4:

[0666] The server integrates facial expression data and voice data. Here, both sets of data are analyzed and the resulting emotional data is structured using Firebase. The output is emotional data formatted for use in decision-making.

[0667] Step 5:

[0668] The server notifies staff members in real time of integrated emotional data via smart glasses. This notification system allows staff to display customer emotional information (the input data) on the smart glasses, enabling immediate responses. The display includes specific customer service advice.

[0669] Step 6:

[0670] Staff members, acting as users, select actions based on customer emotion information displayed on the smart glasses' screen. During interactions with customers, they choose actions to improve service based on the emotion information. This ensures that emotion data, as input, is reflected in practice, leading to improved customer satisfaction.

[0671] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0672] This invention provides a system for accurately and in real time understanding customer emotions in the service industry and improving the quality of service. This system mainly consists of a server, terminals, users, and an emotion engine.

[0673] The server receives video data in real time from cameras installed in the store and applies video processing algorithms to detect visitors' facial expressions. Based on these detected expressions, an emotion classification system categorizes the visitor's emotions. This classification is divided into categories such as joy, surprise, sadness, and dissatisfaction, and the customer's instantaneous emotions are also recorded.

[0674] Simultaneously, the server processes the audio data transmitted from the terminal. The terminal collects audio data through microphones installed in the store and transmits it to the server. The server performs preprocessing such as noise reduction and audio normalization, and estimates the emotion from the audio using an audio emotion estimation method.

[0675] Furthermore, an emotion engine is integrated into the system, enabling multifaceted emotion recognition that includes user behavior data and biometric signals. The emotion engine uses machine learning models to analyze various collected data and recognize the customer's emotional state with high accuracy.

[0676] The server integrates and analyzes facial expression data, voice data, and data from the emotion engine. Based on this analysis, the server generates a report in real time and sends it to the terminal. This report is viewable by administrators and staff and contains customer emotion trends and recommended actions for service improvement.

[0677] For example, if a visitor expresses dissatisfaction during a meal, the camera captures their facial expression, and the server classifies it as "dissatisfaction." Simultaneously, if the word "slow" is extracted from the audio data, the emotion engine, processing this information, recognizes that the user is dissatisfied with the slow service. This information is immediately generated as a report and notified to the staff's terminal. Based on this, the user (staff) can quickly implement countermeasures to improve customer satisfaction. In this way, this system enables real-time emotion recognition and on-the-spot service improvement, resulting in a better customer experience.

[0678] The following describes the processing flow.

[0679] Step 1:

[0680] The server receives video data in real time from cameras installed within the store. The cameras are positioned to capture each table where customers are seated.

[0681] Step 2:

[0682] The server applies a face region detection algorithm to the received video data to identify the faces of individual visitors. A deep learning model is then used to analyze the facial expressions of the identified face regions and classify the emotions.

[0683] Step 3:

[0684] The terminal collects the audio of visitors' conversations through microphones in the store and simultaneously transmits that audio data to a server.

[0685] Step 4:

[0686] The server performs noise reduction and normalization on the received audio data. Next, it applies an audio emotion estimation algorithm to estimate the emotion based on the audio.

[0687] Step 5:

[0688] The server integrates emotional data obtained from facial expressions and emotional data obtained from voice. It then activates an emotion engine and uses a machine learning-based model to accurately recognize the user's emotions from the integrated data.

[0689] Step 6:

[0690] The server analyzes the recognized emotion data to understand customer emotion trends and generates a report based on the analysis results.

[0691] Step 7:

[0692] The terminal receives reports sent from the server and displays them to administrators and staff. Based on these reports, users (staff) take appropriate actions in real time to improve customer satisfaction.

[0693] (Example 2)

[0694] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0695] In the service industry, accurately understanding customer emotions in real time and improving service quality is essential. However, conventional systems have been insufficient in analyzing data obtained from video and audio, making it difficult to understand customer emotions from multiple perspectives. Furthermore, it has been difficult to appropriately utilize the analysis results to improve customer service.

[0696] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0697] In this invention, the server includes video processing means for detecting a visitor's facial expression, emotion classification means for classifying the visitor's emotions based on the facial expression, and acoustic data collection means for acquiring the visitor's voice. This enables multifaceted emotional recognition of the customer.

[0698] "Image processing means" refers to technology for detecting visitors' facial expressions in real time and analyzing the image data.

[0699] "Emotion classification means" refers to a technology that classifies and categorizes a visitor's emotions based on the characteristics of their facial expressions.

[0700] "Acoustic data collection means" refers to technology for collecting sounds generated within a store and pre-processing them into a format suitable for analysis.

[0701] "Voice emotion estimation means" refers to a technology for estimating a visitor's emotions from collected voice data.

[0702] "Emotion recognition means" refers to a technology that analyzes multiple data points, including user behavior data and biosignals, from various angles to recognize visitors' emotions with high accuracy.

[0703] "Data analysis means" refers to technology for integrating the results of facial expression, voice, and emotion recognition and performing detailed analysis.

[0704] "Information generation means" refers to technology for generating output information, including measures to improve customer service, based on analyzed emotional data.

[0705] This invention is configured as a system for understanding customer emotions in the service industry in real time and improving the quality of service. This system consists of multiple elements, including a server, terminals, users, and an emotion engine.

[0706] The server receives real-time video data from high-resolution cameras installed within the store. The server uses video processing algorithms to detect visitors' facial expressions and analyzes this data using emotion classification tools. This allows the server to classify visitors' emotions into categories such as "joy," "surprise," "sadness," and "dissatisfaction."

[0707] The terminal collects audio data through multiple microphones installed within the store and transmits it to a server. The server performs noise reduction and speech normalization on the audio data and estimates the customer's emotions using speech emotion estimation tools. These processes can utilize common open-source speech processing libraries or proprietary speech analysis tools.

[0708] Furthermore, the system incorporates an emotion engine that collects and analyzes multifaceted data, including behavioral data and biosignals. The emotion engine utilizes a generative AI model and machine learning to accurately recognize customer emotions. This allows the server to analyze multiple integrated emotional data sets and generate reports based on the results.

[0709] Users (staff) receive reports notified to their terminals and use the analysis results to improve customer service. For example, if a visitor expresses dissatisfaction in the store, the server analyzes their facial expressions and voice data and creates a report so that staff can respond immediately.

[0710] The following is an example of a specific example and prompt: "If a customer in a restaurant expresses dissatisfaction during their meal, explain how you would analyze their emotions in real time and communicate that information to staff to address the issue."

[0711] The advantage of this system lies in its ability to enable real-time emotion recognition, allowing for quick and effective implementation of countermeasures to improve the customer experience.

[0712] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0713] Step 1:

[0714] The terminal uses cameras installed in the store to acquire real-time video data of visitors. This video data becomes the input to the system. The terminal sends this data directly to the server to prepare for subsequent processing. The actions performed here include camera adjustment and motion detection.

[0715] Step 2:

[0716] The server applies a face detection algorithm to the received video data. This identifies face regions from each frame. The output of this process is the coordinate data of each face region. By identifying the position of the faces, the data necessary for subsequent facial expression analysis is prepared.

[0717] Step 3:

[0718] The server applies an expression recognition algorithm to the data after face detection is complete, classifying emotions. In this process, emotions such as joy, surprise, sadness, and dissatisfaction are determined from facial feature points. The output is a label for the emotion category.

[0719] Step 4:

[0720] The terminal collects customer voice data using microphones installed in the store. This voice data is input to a server, where it undergoes pre-processing such as noise reduction. Since the collected data includes conversation and background noise, it is adjusted to extract important voice components.

[0721] Step 5:

[0722] The server receives denoised audio data and applies an audio emotion estimation algorithm to it. Based on the features extracted from the audio, it estimates the customer's emotions. The output is an emotion label from the audio, ready to be integrated with facial expression data.

[0723] Step 6:

[0724] The server integrates emotional information obtained from both facial expression and voice data into an emotion engine. This emotion engine uses a generative AI model and machine learning to perform comprehensive emotional analysis. The analysis results of multiple emotional data are output and transformed into a form that is meaningful to the user.

[0725] Step 7:

[0726] The server generates a report based on the analyzed sentiment data. The output includes customer sentiment trends and suggestions for service improvement. This report is sent to terminals, allowing users (staff) to understand it in real time and use it to improve their responses.

[0727] (Application Example 2)

[0728] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0729] In traditional service industries, it is difficult to instantly grasp and respond to customer emotions, limiting the potential for improving service quality. There is a need for technology that can analyze customer emotions in real time and enable service providers to respond appropriately on the spot.

[0730] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0731] In this invention, the server includes a video generation means for detecting a visitor's facial expression, an emotion classification means for classifying the visitor's emotions based on the facial expression, and an audio acquisition means for acquiring the visitor's sound. This allows service staff to recognize customers' emotions in real time and immediately improve their service.

[0732] The "image generation means" is a mechanism that acquires and processes the video data necessary to detect the facial expressions of visitors.

[0733] An "emotion classification system" is a mechanism that classifies a visitor's emotions based on acquired facial expression data.

[0734] "Sound acquisition means" refers to a mechanism for acquiring sound data of visitors.

[0735] An "acoustic emotion estimation means" is a mechanism that estimates emotions based on acoustic data.

[0736] "Information analysis means" refers to a mechanism that integrates and analyzes emotional data based on facial expressions and sounds.

[0737] An "information generation means" is a mechanism that generates output information using analyzed emotion data.

[0738] A "wearable display device" is a device equipped with a display device for presenting output information to customers.

[0739] This invention is a system that grasps customer emotions in real time in physical stores and improves the quality of service. The server processes video data acquired from cameras installed in the store using video generation means for detecting visitors' facial expressions and performs facial recognition. Furthermore, it collects acoustic data from microphones installed in the store using acoustic acquisition means for acquiring acoustic data of visitors and estimates acoustic emotions.

[0740] The server uses information analysis tools to comprehensively analyze this data and recognize the customer's emotional state with high accuracy. From the analyzed emotional data, an information generation tool generates output information, which is then presented to the service provider on the spot via a wearable display device. This information provides crucial clues for the service provider to take quick action.

[0741] Specific hardware includes wearable display devices such as smart glasses. Software used includes OpenCV for video processing, librosa for audio processing, and scikit-learn for machine learning models. This enables real-time customer emotion recognition and rapid service improvement. A concrete application of the invention is that, for example, if a customer shows dissatisfaction while waiting, their facial expression and sound can be used to recognize their emotion as "dissatisfaction," allowing staff to respond quickly and improve customer satisfaction.

[0742] An example of a specific prompt for the generating AI model is, "Based on the visitor's facial expressions and audio data, please list the most suitable recommended actions to improve customer service on the spot." This allows service staff to quickly implement appropriate countermeasures.

[0743] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0744] Step 1:

[0745] The server acquires video data from cameras installed within the store. The input is real-time video data, and the output is image data including the facial regions of visitors. The video generation method uses OpenCV to perform face recognition and extracts facial regions of particular interest.

[0746] Step 2:

[0747] The server acquires acoustic data from microphones installed within the store. The input is acoustic data including ambient noise, and the output is pre-processed acoustic data. The acoustic acquisition method uses librosa to remove noise and normalize the acoustic signal.

[0748] Step 3:

[0749] The server classifies emotions based on facial expression data. The input is image data of the face region, and the output is the classified emotion category (e.g., joy, surprise, dissatisfaction). The emotion classification means analyzes facial landmarks and estimates emotions using a machine learning model.

[0750] Step 4:

[0751] The server estimates acoustic emotion based on acoustic data. The input is normalized acoustic data, and the output is the estimated emotion category. The acoustic emotion estimation means performs emotion estimation based on voice tone and phoneme information.

[0752] Step 5:

[0753] The server integrates and analyzes emotional information obtained from facial expression data and acoustic data. The input is multiple emotional data, and the output is the overall emotional state of the customer. The information analysis means integrates different emotional indicators with weights to generate an overall picture.

[0754] Step 6:

[0755] The server generates output information based on the analysis results. The input is the overall emotional state, and the output is a report containing recommended actions for improving customer service. The information generation means uses a generation AI model to create a list based on prompt sentences.

[0756] Step 7:

[0757] The user receives output information generated through a wearable display device. The input is a real-time report, and the output manifests as a rapid action taken by the service provider. Display devices such as smart glasses visually notify the service provider of the emotional state.

[0758] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0759] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0760] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0761] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0762] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0763] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0764] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0765] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0766] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0767] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0768] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0769] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0770] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0771] 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.

[0772] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0773] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0774] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0775] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0776] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0777] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0778] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0779] The following is further disclosed regarding the embodiments described above.

[0780] (Claim 1)

[0781] An image generation means for detecting the facial expressions of visitors,

[0782] An emotion classification means for classifying the emotions of visitors based on the aforementioned facial expressions,

[0783] A means of acquiring audio for obtaining the voice of a visitor,

[0784] A voice emotion estimation means for estimating emotions based on the aforementioned voice,

[0785] A data analysis means for integrating and analyzing the aforementioned emotional data based on facial expressions and voice,

[0786] A system including a report generation means for generating output information using the analyzed emotion data.

[0787] (Claim 2)

[0788] The system according to claim 1, wherein the voice acquisition means acquires voice data within a specific range and performs preprocessing.

[0789] (Claim 3)

[0790] The system according to claim 1, wherein the image generation means processes video data captured in real time and recognizes individual face regions.

[0791] "Example 1"

[0792] (Claim 1)

[0793] Image processing means for detecting the facial expressions of visitors,

[0794] A data classification means for classifying the emotions of visitors based on the information extracted from the aforementioned facial expressions,

[0795] A voice processing means for acquiring and processing the voice of a visitor,

[0796] An acoustic analysis means for estimating emotions based on the aforementioned characteristics of the voice,

[0797] Information analysis means for integrating and analyzing the data from facial expressions and voice,

[0798] A system including information generation means for generating report information using the analyzed emotion data.

[0799] (Claim 2)

[0800] The system according to claim 1, wherein the audio processing means acquires an audio signal within a predetermined range and performs preliminary processing.

[0801] (Claim 3)

[0802] The system according to claim 1, wherein the image processing means processes video information acquired in real time and recognizes individual face regions.

[0803] "Application Example 1"

[0804] (Claim 1)

[0805] An image generation means for detecting the facial expressions of visitors,

[0806] An emotion classification means for classifying the emotions of visitors based on the aforementioned facial expressions,

[0807] A means of acquiring audio for obtaining the voice of a visitor,

[0808] A voice emotion estimation means for estimating emotions based on the aforementioned voice,

[0809] A data analysis means for integrating and analyzing the aforementioned emotional data based on facial expressions and voice,

[0810] A report generation means for generating output information using the analyzed emotion data,

[0811] A notification means for notifying an information display device in real time of the analyzed emotion data, enabling immediate response,

[0812] Display means for displaying the aforementioned emotional data on individual devices that interact with visitors, enabling users to facilitate services based on the visitor's state.

[0813] A system that includes this.

[0814] (Claim 2)

[0815] The system according to claim 1, wherein the voice acquisition means acquires voice data within a specific range and performs preprocessing, and estimates emotions based on the characteristics of the voice.

[0816] (Claim 3)

[0817] The system according to claim 1, wherein the image generation means processes video data captured in real time and not only recognizes individual face regions but also updates emotional information in response to changes in facial expression.

[0818] "Example 2 of combining an emotion engine"

[0819] (Claim 1)

[0820] A video processing means for detecting the facial expressions of visitors,

[0821] An emotion classification means for classifying the emotions of visitors based on the aforementioned facial expressions,

[0822] A means for collecting acoustic data to obtain the voices of visitors,

[0823] A voice emotion estimation means for estimating emotions based on the aforementioned voice,

[0824] An emotion recognition means for analyzing multiple data, including user behavior and biosignals, from various perspectives,

[0825] A data analysis means for integrating and analyzing emotional data based on the results of facial expression, voice, and emotion recognition,

[0826] A system including information generation means for generating output information, including improvement measures, using the analyzed emotion data.

[0827] (Claim 2)

[0828] The system according to claim 1, wherein the acoustic data acquisition means acquires and preprocesses audio data within a specific range.

[0829] (Claim 3)

[0830] The system according to claim 1, wherein the image processing means processes image data acquired in real time and recognizes individual face regions.

[0831] "Application example 2 when combining with an emotional engine"

[0832] (Claim 1)

[0833] A means for generating images to detect the facial expressions of visitors,

[0834] An emotion classification means for classifying the emotions of visitors based on the aforementioned facial expressions,

[0835] A means of acquiring sound for obtaining the sound of visitors,

[0836] Acoustic emotion estimation means for estimating emotions based on the aforementioned sound,

[0837] Information analysis means for integrating and analyzing the aforementioned emotional data based on facial expressions and sounds,

[0838] Information generation means for generating output information using the analyzed emotion data,

[0839] A system including a wearable display device equipped with a display device that presents output information to customer service personnel.

[0840] (Claim 2)

[0841] The system according to claim 1, wherein the sound acquisition means acquires sound data within a specific range and performs preprocessing.

[0842] (Claim 3)

[0843] The system according to claim 1, wherein the video generation means processes video data captured in real time and recognizes individual facial regions, and includes a wearable display device for notifying the customer of the emotional result. [Explanation of Symbols]

[0844] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. An image generation means for detecting the facial expressions of visitors, An emotion classification means for classifying the emotions of visitors based on the aforementioned facial expressions, A means of acquiring audio for obtaining the voice of a visitor, A voice emotion estimation means for estimating emotions based on the aforementioned voice, A data analysis means for integrating and analyzing the aforementioned emotional data based on facial expressions and voice, A system including a report generation means for generating output information using the analyzed emotion data.

2. The system according to claim 1, wherein the voice acquisition means acquires voice data within a specific range and performs preprocessing.

3. The system according to claim 1, wherein the image generation means processes video data captured in real time and recognizes individual face regions.

Images