system

Abstract

We provide the system. [Solution] A means of receiving video data and analyzing it to identify the movements of a person, A means of acquiring audio data and converting it into text data, A means for detecting abnormal behavior based on analyzed video and audio data, A means for generating and notifying a warning report when abnormal behavior is detected, A system that includes this.

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In a modern workplace environment, especially in a closed space such as a conference room, there is a problem that harassment is likely to occur and its detection is difficult. There is a need for an appropriate system to ensure the safety and well-being of employees, improve productivity, and foster a healthy workplace culture. Also, it is necessary to improve the efficiency of the entire organization by smoothly progressing meetings and accurately evaluating the contribution degrees of participants.

Means for Solving the Problems

[0005] This invention provides a system equipped with means for receiving and analyzing video data to identify human movements, and by utilizing means for acquiring audio data and converting it into text data, it offers accurate information for detecting abnormal behavior. Furthermore, when abnormal behavior is detected, it has a function to generate a warning report and immediately notify, contributing to the early resolution of problems. In addition, it promotes smooth communication by monitoring the progress of meetings, detecting points where discussions stall, and automatically generating and presenting suggestions. It also supports the growth of each employee by evaluating their job performance and generating feedback reports.

[0006] "Video data" refers to visual information acquired from video recording devices such as cameras, represented in digital format.

[0007] "Means for identifying actions" refers to technical methods or devices that analyze video data to identify the movements and postures of a subject and classify them into specific actions.

[0008] "Audio data" refers to digitalized sound information collected through devices such as microphones.

[0009] "Text data" refers to a data format obtained by converting audio data into text information using speech recognition technology.

[0010] "Abnormal behavior" refers to actions that deviate from predetermined normal behavioral patterns, and specifically includes behaviors such as harassment and inappropriate conduct.

[0011] A "warning report" is a report or notification containing details of a problem, generated when the system detects abnormal behavior, and is intended to be promptly communicated to the relevant parties.

[0012] "Monitoring the progress of a meeting" refers to the act or technology of observing in real time whether a meeting is proceeding smoothly and detecting stagnation or inefficiencies in the discussion.

[0013] "Generating and presenting proposals" refers to creating suggestions, such as those using AI, to guide the next action steps or direction of discussion, based on the meeting's situation, and then presenting these suggestions to the meeting participants.

[0014] A "feedback report" is a report that summarizes the results of an evaluation of an individual employee's job performance and behavior, and includes areas for improvement and points of commendation. [Brief explanation of the drawing]

[0015] [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]It is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] It is a sequence diagram showing the processing flow of the data processing system in Embodiment 2 when the emotion engine is combined. [Figure 14] It is a sequence diagram showing the processing flow of the data processing system in Application Example 2 when the emotion engine is combined.

Modes for Carrying Out the Invention

[0016] Hereinafter, an example of an embodiment of the system according to the technology of the present disclosure will be described with reference to the accompanying drawings.

[0017] First, the terms used in the following description will be explained.

[0018] In the following embodiments, the numbered processor (hereinafter simply referred to as "processor") may be one arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be one type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), etc.

[0019] In the following embodiments, the numbered RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0020] In the following embodiments, the signed storage is one or more non-volatile storage devices that store various programs and various parameters. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes.

[0021] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0022] 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."

[0023] [First Embodiment]

[0024] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0033] 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.

[0034] 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.

[0035] 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".

[0036] This invention relates to a system that uses devices such as cameras and microphones to analyze human movements and voices in real time within a specific space, such as a workplace or meeting room, and detects potentially harassing behavior. This system utilizes both video and audio data to identify abnormal behavior, generate warning reports, and aims to maintain workplace safety and well-being.

[0037] Video and audio data analysis

[0038] The server receives video data from the camera and performs video analysis using machine learning. This analysis captures the movements and facial expressions of people in the video in real time and detects abnormal behavior based on specific patterns. The server also acquires audio data from microphones in the conference room and converts it into text using speech recognition software. Natural language processing techniques are then applied to this text data to identify aggressive remarks and negative emotional expressions.

[0039] Detection and reporting of abnormal behavior

[0040] When an anomaly is detected, the server immediately generates an alert report. This report contains detailed information about the abnormal behavior and is sent to the designated administrator or person in charge. For example, if a particular employee uses inappropriate language towards other participants during a meeting, the system can detect this and send a warning message to the administrator, enabling a quick response to the problem.

[0041] Monitoring and suggesting meeting progress

[0042] During meetings, the server monitors the progress of the meeting and detects situations where the discussion has stalled or a particular topic is not progressing smoothly. In such cases, the server uses AI to generate appropriate suggestions and presents them to meeting participants via their terminals. For example, if the discussion reaches an impasse, participants may be offered a suggestion such as, "What are other opinions on this topic?" to help the meeting proceed smoothly.

[0043] Feedback and evaluation

[0044] Furthermore, the system's server records and evaluates each employee's contributions and job performance during meetings. Based on this evaluation, a feedback report is generated and delivered to individual employees via their terminals. This report specifically details areas for improvement and commendable aspects, allowing each employee to use it for self-improvement.

[0045] With the configuration described above, the present invention simultaneously achieves the optimization of the workplace environment, improved employee productivity, and the cultivation of a healthy organizational culture.

[0046] The following describes the processing flow.

[0047] Step 1:

[0048] The server receives video data in real time from cameras installed in the workplace. This prepares it for continuous monitoring of people and situations within the target area.

[0049] Step 2:

[0050] The server activates a machine learning module to analyze the received video data, examining the movements and facial expressions of individuals frame by frame. This allows it to identify specific behavioral patterns and assess the likelihood of abnormal behavior.

[0051] Step 3:

[0052] The server acquires audio data from microphones in the conference room. The acquired audio data is converted into a digital format and then converted into text data by speech recognition software.

[0053] Step 4:

[0054] The server applies natural language processing algorithms to the converted text data to analyze the tone and content of the speech. This identifies offensive words and negative expressions.

[0055] Step 5:

[0056] If abnormal behavior or inappropriate remarks are detected, the server automatically generates a warning report based on that information. This report is immediately notified to administrators and responsible personnel.

[0057] Step 6:

[0058] During the meeting, the server monitors the progress and determines if the discussion is stalled or if information is lacking. If necessary, the AI ​​generates suggestions and presents solutions for improvement.

[0059] Step 7:

[0060] The generated suggestions are displayed to meeting participants via their devices, helping to ensure the meeting runs more smoothly.

[0061] Step 8:

[0062] After the meeting, the server evaluates each employee's contribution and generates a feedback report based on that data. The report is distributed to each employee via their terminal, which helps them with individual analysis and improvement.

[0063] Through this series of processes, the system helps optimize the workplace environment, prevent harassment, improve meeting efficiency, and promote employee growth.

[0064] (Example 1)

[0065] 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."

[0066] Detecting inappropriate behavior and remarks in the workplace and meetings, and responding promptly to them, has long been a challenge. Furthermore, there is a need for a system that efficiently monitors meeting progress, makes appropriate suggestions, evaluates employee contributions, and provides feedback. Improving these aspects is essential to enhancing workplace safety and productivity, and fostering a healthy organizational culture.

[0067] 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.

[0068] In this invention, the server includes means for receiving and analyzing video information, means for acquiring audio information and converting it into text information, and means for detecting abnormal behavior and generating warning reports based on the analyzed information. This makes it possible to detect inappropriate behavior in real time in the workplace or meetings and respond quickly. Furthermore, by providing means for monitoring discussions and making suggestions using AI generation, and means for evaluating job performance and generating feedback reports, it appropriately supports the efficiency of meetings and the individual performance evaluation of employees.

[0069] "Visual information" refers to visual data collected by cameras and other recording devices.

[0070] "Audio information" refers to sound data collected using microphones or other recording devices.

[0071] "Analysis" is the process of analyzing information and extracting meaningful patterns and features contained within it.

[0072] "Abnormal behavior" refers to actions or words that are unusual or considered particularly problematic.

[0073] A "warning report" is a report generated when abnormal behavior is detected, containing details of the situation and information about the need for action.

[0074] "Progress of discussion" refers to the current state of a meeting or discussion, indicating its progress and the degree of development of its content.

[0075] "Generative AI" refers to artificial intelligence systems that automatically generate data and use that data to make appropriate suggestions and decisions.

[0076] A "feedback report" is a document or information that includes an evaluation of a specific activity or outcome, as well as suggestions for improvement.

[0077] The present invention provides real-time detection of harassment and abnormal behavior in the workplace and meeting rooms, support for meeting progress, and employee evaluation and feedback based on video and audio information. This system primarily relies on a server to operate, combining various devices and technologies.

[0078] The server continuously collects video and audio information through cameras and microphones. This allows the server to monitor actions and speech in real time. High-resolution cameras and machine learning software are used to analyze the video information, for example, using OpenCV or TENSORFLOW® to recognize actions. Face recognition and behavior recognition identify a person's actions and facial expressions.

[0079] Audio information is converted to text using speech recognition technology on the server, and then natural language processing is performed. This process uses standard speech recognition software or natural language processing libraries such as spaCy to identify aggressive remarks and negative emotional expressions. As a result, if abnormal behavior is detected, a warning report is generated immediately and notified to administrators or responsible personnel.

[0080] Furthermore, the server monitors the meeting's progress using an AI model and generates appropriate suggestions when the discussion stalls. This includes a feature that uses a generative AI model to provide participants with suggestions to facilitate conversation, such as, "What are other opinions on this topic?" This is expected to ensure the meeting runs smoothly and elicit more opinions.

[0081] The server also records each employee's contributions to meetings and their work performance, and evaluates them based on this data. Based on these evaluations, a feedback report is generated and delivered individually to each employee via their terminal. This report includes specific areas for improvement and points of praise, which users can use to improve themselves.

[0082] As described above, this invention is a system that improves the quality of communication in the workplace and meetings, and supports the prevention of harassment and the efficiency of job evaluations.

[0083] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0084] Step 1:

[0085] The server continuously acquires video and audio information from cameras and microphones. The input is on-site video and audio, which is received in stream format. Specifically, it collects video data from high-resolution cameras and audio data from high-sensitivity microphones. The output is this video and audio data.

[0086] Step 2:

[0087] The server analyzes the acquired video information. The input is the video data acquired in step 1. A machine learning model is used to identify the actions and facial expressions of people in each frame. Specifically, image processing is performed using libraries to perform face recognition and gesture recognition. The output is the identified people and their action patterns.

[0088] Step 3:

[0089] The server converts the audio information into text. The input is the audio data obtained in step 1. Speech recognition technology is used to transcribe the audio into text, preparing it for application to natural language processing technology. Standard speech recognition software is used for this process. The output is the text data corresponding to the audio data.

[0090] Step 4:

[0091] The server detects aggressive remarks from the analyzed speech-to-text. The input is the text data obtained in step 3. Natural language processing techniques are used to analyze the tone and content of the words and identify abnormal remarks. Specific sentiment analysis methods are used in this process. The output is the identified aggressive words and their context.

[0092] Step 5:

[0093] The server generates a warning report based on abnormal behavior or statements and notifies the administrator. The input is the abnormal behavior or statements detected in steps 2 and 4. The warning report, including a detailed description of the situation, is sent to the relevant administrator via email or the notification system. The output is the generated warning report.

[0094] Step 6:

[0095] The server monitors the progress of the meeting and uses a generative AI model to create appropriate suggestions. The input is meeting progress data. If the discussion stalls, the AI ​​is used to generate suggestions such as "offer an opinion from a different perspective" and provides them to participants via their terminals. The output is suggestions for facilitating the smooth progress of the meeting.

[0096] Step 7:

[0097] The server evaluates employee contributions and performance and generates individual feedback reports. Input data consists of statements and actions during meetings. Based on this data, the server performs an evaluation and creates a detailed report outlining areas for improvement and areas for praise for each employee, which is then sent to each employee individually via their terminal. The output is a detailed feedback report.

[0098] (Application Example 1)

[0099] 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."

[0100] Detecting harassment and abnormal behavior in the workplace often relies on manual processes and human monitoring, making efficient responses difficult. Furthermore, decreased productivity due to work stagnation and communication breakdowns is a serious problem. Traditional methods cannot provide rapid and effective solutions to these issues. Therefore, there is a need to maintain a safe and healthy workplace environment and improve the quality of work.

[0101] 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.

[0102] In this invention, the server includes means for receiving and analyzing video information to identify human movements, means for acquiring audio information and converting it into text information, and means for detecting abnormal behavior based on the analyzed video and audio information. This enables real-time detection of abnormal behavior in the workplace, rapid notification of warning reports, and visualization of abnormal behavior using an augmented reality display device.

[0103] "Visual information" refers to visual data acquired by devices such as cameras and sensors.

[0104] "Audio information" refers to acoustic data collected by devices such as microphones.

[0105] "Text information" refers to information that is expressed in written form based on audio information.

[0106] "Analysis" refers to the process of using acquired data to analyze it according to a specific purpose.

[0107] "Abnormal behavior" refers to behavior that deviates from normal behavioral patterns and is considered problematic based on specific criteria.

[0108] A "warning report" refers to the notification content generated when abnormal behavior is detected, and includes information to prompt appropriate action.

[0109] An "augmented reality display device" refers to a device equipped with technology that overlays digital information onto the real world's field of view.

[0110] The system for implementing this invention consists of software and hardware that enables both video and audio processing. The server receives video and audio information from sensor devices such as cameras and microphones. The video information is analyzed in real time using video analysis libraries such as OpenCV to identify human movements. The audio information is converted into text using speech recognition software such as DeepSpeech, and then analyzed using natural language processing techniques such as NLTK. This process enables the rapid detection of abnormal behavior.

[0111] If abnormal behavior is detected, the server generates a warning report and sends it to an augmented reality display device for notification. This allows users present to visually receive information about the abnormal behavior. For example, if an employee speaks in a harsh tone at work, the system detects the audio information and displays a warning message such as "Strong language is being used" on the augmented reality display device to alert the user.

[0112] As a concrete example, focusing on the theme of "real-time detection and visualization of abnormal behavior in workplace meetings," the system analyzes video and audio to monitor the attitudes and statements of meeting participants. Using this system, users can quickly respond to subtle nuances that were previously difficult to detect. When utilizing a generative AI model, an appropriate prompt would be, "Design an optimal algorithm to analyze real-time video and audio data from the workplace and report abnormal behavior and incidents constituting harassment."

[0113] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0114] Step 1:

[0115] The server receives video and audio information acquired from the terminal via the camera and microphone. Since this input data is processed in real time, it is necessary to maintain a smooth data flow.

[0116] Step 2:

[0117] The server uses the OpenCV library to analyze video information. The purpose of the analysis is to identify human movements and facial expressions, and to extract data to identify specific abnormal behavioral patterns. The input is video information, and the output is the analyzed behavioral patterns.

[0118] Step 3:

[0119] The server uses DeepSpeech to convert audio information into text information. This conversion process extracts highly accurate text data from the audio, and sentiment analysis is performed based on this data. The input is audio information, and the output is text information.

[0120] Step 4:

[0121] The server uses NLTK to perform natural language processing on text information. Specifically, it detects aggressive remarks and negative emotional expressions in the text and flags them as abnormal behavior. The input is the transformed text information, and the output is the analysis results and the abnormal behavior flag.

[0122] Step 5:

[0123] If abnormal behavior is detected, the server generates a warning report. This report includes specific details of the detected abnormal behavior and related information. The generated report is sent to an augmented reality display device for visualization. The input is the abnormal behavior flag, and the output is the warning report.

[0124] Step 6:

[0125] The user receives a visually represented warning report through an augmented reality display device. The specific details of abnormal behavior are provided visually, serving as a tool to prompt immediate action. The input is the warning report, and the output is the user's recognition and response.

[0126] 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.

[0127] This invention is a system that detects abnormal behavior and takes appropriate action by analyzing the actions and emotions of individuals in the workplace in real time. In addition to analyzing video and audio data, this system combines an emotion engine to identify the user's emotions, contributing to improved workplace safety and employee well-being.

[0128] Video and audio data analysis

[0129] The server acquires video data from cameras placed in the workplace area and analyzes the behavior using machine learning algorithms. This makes it possible to monitor the behavior of specific individuals in real time and detect signs of abnormal behavior. The server also collects audio data from microphones and converts it into text data. Based on this text data, natural language processing technology is used to identify linguistic anomalies.

[0130] Emotion recognition by an emotion engine

[0131] The server uses an emotion engine to analyze the user's emotional state from video and audio data. The emotion engine detects facial expressions and tone of voice to determine emotions such as joy, anger, and sadness. This enables more accurate detection of abnormal behavior by incorporating emotional data.

[0132] Detection and response to abnormal behavior

[0133] If abnormal behavior or inappropriate emotional states are detected, the server immediately generates an alert report and notifies the relevant administrator or person in charge. This report records details of the behavior and emotions, helping to quickly determine appropriate action. For example, if it is detected that a participant is emotionally agitated and showing discomfort during a meeting, the situation can be reported to the administrator, who can then be instructed to take appropriate action to prevent the problem from escalating.

[0134] Post-meeting evaluation and feedback

[0135] After the meeting ends, the server analyzes data including each participant's behavior and emotions to create individual feedback reports. These reports, delivered to employees via their devices, include trends and areas for behavioral improvement based on their emotional state, promoting self-improvement.

[0136] In this way, by utilizing information about emotional states, the system can simultaneously improve employees' emotional well-being and job performance.

[0137] The following describes the processing flow.

[0138] Step 1:

[0139] The server receives video data in real time from cameras installed in the workplace and sends the data to a sequential processing stream. This prepares the data for real-time analysis.

[0140] Step 2:

[0141] The server applies machine learning models to the received video data to analyze the movements and facial expressions of individual people. It uses face recognition and motion detection algorithms to extract deviations from normal behavioral patterns.

[0142] Step 3:

[0143] The server acquires audio data from microphones installed in the conference room or designated area and stores it as digital audio. The audio processing module then passes this to the speech recognition engine, which converts it into text data.

[0144] Step 4:

[0145] The server analyzes the converted text data using natural language processing algorithms, identifying aggressive or suspicious behavior by analyzing the content and tone of the statements.

[0146] Step 5:

[0147] The server uses an emotion engine to identify emotional states from video and audio data. By analyzing changes in facial features and voice pitch, it classifies the user's emotions into categories such as "joy," "anger," and "sadness."

[0148] Step 6:

[0149] If abnormal behavior or negative emotional states are detected, the server generates an alert report summarizing the situation. This report includes detailed observations and information necessary for action, and is immediately sent to the administrator.

[0150] Step 7:

[0151] The terminal displays suggestions generated during the meeting to participants. These suggestions include specific measures to stimulate discussion based on sentiment data, thereby improving the quality of the meeting.

[0152] Step 8:

[0153] After the meeting, the server analyzes each participant's behavioral and emotional data to perform individual evaluations and generate feedback reports. These reports are distributed to employees via their devices to support their individual growth.

[0154] Step 9:

[0155] Based on the feedback received, users can develop action plans for self-improvement. The feedback serves as a guide for specific behavioral modifications.

[0156] Through these steps, the system comprehensively monitors the work environment, enabling early detection of anomalies and simultaneous improvement of productivity.

[0157] (Example 2)

[0158] 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".

[0159] In the workplace, effectively detecting and promptly responding to abnormal employee behavior and emotional changes is crucial for improving safety and operational efficiency. However, conventional systems are limited to detection based on behavioral and verbal abnormalities, making it difficult to achieve highly accurate detection and immediate response that takes emotional states into account.

[0160] 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.

[0161] In this invention, the server includes means for receiving and analyzing video data to identify a person's movements, means for acquiring audio data and converting it into text data, and means for identifying emotional states from the analyzed data. This enables highly accurate detection of emotional changes and abnormal emotions, and allows for rapid warning reporting and feedback.

[0162] "Video data" refers to visual information acquired from cameras in the workplace environment, and serves as basic data for analyzing people's movements and facial expressions.

[0163] "Means for identifying movements" refers to technologies that analyze video data to identify the movements and behavioral patterns of individual people and detect abnormal movements.

[0164] "Audio data" refers to acoustic signals acquired through a microphone, and is used to analyze conversation content and voice tone.

[0165] "Methods for converting to text data" refers to methods for converting audio data into text data, making the content analyzable.

[0166] A "means for detecting abnormal behavior" refers to a system that identifies unusual behavior or emotional changes based on video and audio data.

[0167] "Means for generating alert reports" refers to a process for documenting details of abnormal behavior or emotions when they are detected and notifying relevant parties.

[0168] "Means for identifying emotional states" refers to technologies that analyze video and audio data to determine a person's emotions (e.g., joy, anger, sadness).

[0169] "Methods for providing feedback" refer to techniques used to show participants areas for improvement and emotional tendencies based on analyzed behavioral and emotional data.

[0170] This invention is a system that analyzes human behavior and emotions in real time and detects anomalies in order to improve safety and employee well-being in the workplace environment. The main components of the invention consist of a camera, microphone, server, and terminal.

[0171] The server processes video data collected by the camera and uses TensorFlow to analyze behavior using machine learning algorithms, including open-source libraries. Through this analysis, the server identifies the behavioral state of specific individuals and detects abnormal behavior. Additionally, audio data acquired by the microphone is converted into text using speech recognition services such as Google Cloud Speech-to-Text, and natural language processing techniques are used to identify linguistic anomalies.

[0172] Furthermore, the server uses an emotion recognition API to analyze emotional states from video and audio. This analysis can detect facial expressions and voice tone, and identify emotions such as joy or anger. The emotion data improves the accuracy of detecting abnormal behavior.

[0173] When an anomaly is detected, an alert report is immediately generated and notified to the appropriate administrator. This report records specific details of the behavior and emotions, facilitating a swift response. For example, if a participant becomes emotionally unstable during a meeting, the system can detect this change and notify the administrator, preventing the problem from escalating.

[0174] Furthermore, after the meeting, the server comprehensively analyzes the participants' behavior and emotions and generates a feedback report. This report includes patterns of emotional changes and points for improving behavior, and is provided to employees via their terminals. This allows users to understand their own emotional tendencies and aim to improve their work performance.

[0175] Examples of prompt messages for this system include the following:

[0176] "Please tell me how to analyze emotional data in the workplace to detect abnormal behavior and suggest improvements."

[0177] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0178] Step 1:

[0179] The server acquires video and audio data in real time from cameras and microphones placed in the workplace. It receives video and audio data streams as input and prepares them as datasets for processing. Specific operations include temporal slicing and buffering of the data.

[0180] Step 2:

[0181] The server uses the acquired video data to analyze human movement by applying machine learning models such as TensorFlow. It uses video frames as input and decodes the action in each frame using a motion recognition algorithm. This process generates action labels as output, enabling the detection of anomalies in the movement of people and objects.

[0182] Step 3:

[0183] The server converts audio data into text data using a speech recognition service such as Google Cloud Speech-to-Text. It takes the audio signal as input, extracts acoustic features, and sequentially converts the audio into text information. The output of this process is conversation data in text format.

[0184] Step 4:

[0185] The server analyzes the converted text data using natural language processing techniques to identify linguistic anomalies. The input is text data, and semantic analysis is performed via a natural language processing engine to detect abnormal language patterns. The output includes alert information for unusual language usage.

[0186] Step 5:

[0187] The server uses an emotion recognition API to perform emotion analysis based on the analyzed video and audio. It takes image features and audio spectra as input and uses a deep learning model to classify emotion labels. The output is generated as a dataset of emotional states, identifying emotions such as joy, anger, and sadness.

[0188] Step 6:

[0189] When abnormal behavior or emotional changes are detected, the server generates an alert report and notifies administrators and other relevant parties who require attention. At this stage, the anomaly detection data is used as input, and a detailed anomaly report is created using a template for structuring the report. The generated output is sent via email or message notification.

[0190] Step 7:

[0191] After the event ends, the server analyzes data for each participant and generates feedback reports. Using behavioral and emotional datasets as input, statistical models and evaluation algorithms are employed to generate reports on emotional tendencies and areas for improvement for each employee. The final output is the report delivered via the terminal.

[0192] This processing flow enables the system to achieve real-time safety in the workplace and improve employee well-being.

[0193] (Application Example 2)

[0194] 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".

[0195] In the workplace, it is crucial to manage worker safety and mental health simultaneously and to detect abnormal behavior and emotional changes early. However, conventional systems lack the means to comprehensively analyze and notify this information in real time, resulting in a challenge in adequately addressing the improvement of workplace safety and worker well-being.

[0196] 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.

[0197] In this invention, the server includes means for receiving and analyzing video data to identify the actions of an individual, means for acquiring audio data and converting it into text data, means for detecting abnormal behavior based on the analyzed video and audio data, means for monitoring the individual's biological information in real time and analyzing their emotional state, and means for sending an alarm to a management device and instructing appropriate action based on the emotional analysis. This enables real-time monitoring of changes in the actions and emotions of workers in the workplace environment, allowing for a rapid response.

[0198] "Video data" refers to digital or analog data representing visual information acquired by a camera or other imaging device.

[0199] "Analysis" is a method of processing collected data to extract specific information or understand its characteristics.

[0200] An "individual" is an object being monitored within a system, usually referring to a human worker, but may also include other living organisms or objects.

[0201] "Action" refers to the physical actions or movements performed by an individual, and usually means bodily movements.

[0202] "Audio data" refers to sound information acquired through devices such as microphones, expressed in digital or analog format.

[0203] "Converting to text data" is the process of analyzing information such as audio and video and converting it into digital format as text.

[0204] "Detection of abnormal behavior" refers to identifying behaviors that deviate from normal behavioral patterns and recognizing the signs of such behavior.

[0205] "Biometric information" refers to data about an individual's physical state and function, including heart rate and respiratory rate.

[0206] "Analysis of emotional state" is a process that aims to evaluate an individual's psychological state and identify specific emotions such as joy or anger.

[0207] "Sending an alarm" is a systemic action to notify relevant parties of information when an anomaly or specific conditions are met.

[0208] "Management device" refers to hardware or software that controls the entire monitoring system and aggregates or distributes necessary information.

[0209] To implement this invention, a system is required in which multiple devices and software components work in cooperation. The server simultaneously acquires video and audio data from cameras and microphones in the workplace environment and analyzes them to identify the actions and emotional states of individuals. Specifically, the video data is processed in real time using an image processing library such as OpenCV, and an emotion recognition engine called EmotionEngine determines emotions from the individual's facial expressions. The audio data is converted into text data using an audio processing library such as librosa.

[0210] If the analysis detects abnormal behavior or emotional abnormalities such as stress, the server sends an alert to the management device, which then notifies the administrator's smartphone or computer. This management device uses a module called AlertManager to generate the alert. Appropriate countermeasures are automatically suggested, enabling a rapid response to the situation.

[0211] As a concrete example, when a new production line starts operation, the system automatically monitors whether employees are experiencing stress due to the change in environment. If an anomaly is detected, managers can use that information to readjust the work environment settings and implement measures to reduce the burden on workers.

[0212] To realize such a system, it is possible to obtain more detailed analysis results by supplying advanced prompt sentences to the generating AI model. For example, a prompt sentence such as, "Please give me ideas for an effective monitoring system to analyze abnormal behavior and emotions in the work environment," could be considered.

[0213] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0214] Step 1:

[0215] The server acquires video data from cameras installed in the workplace environment. It analyzes the received video data using OpenCV to identify individual movements. Specifically, it extracts the coordinates and movement features of individuals and passes this data to subsequent processing steps.

[0216] Step 2:

[0217] The server acquires audio data from microphones in the environment. This audio data is converted into text data using the librosa library. The audio waveform, as input, is analyzed into frequency components, and the content is output in text format through a speech recognition algorithm. This text data is used to detect abnormal behavior.

[0218] Step 3:

[0219] The device uses EmotionEngine to recognize emotional states from analyzed video and audio data. It analyzes facial expressions and tone of voice as input and outputs emotions such as anger and joy. This allows for the identification of signs of abnormal behavior based on emotions.

[0220] Step 4:

[0221] The server detects abnormal behavior based on analyzed behavioral and emotional data. If an abnormality is detected, it sends an alert to the management device using AlertManager. By analyzing the input abnormality information and immediately generating an alert and notifying the administrator's terminal, it encourages a quick response.

[0222] Step 5:

[0223] After receiving an alert as an administrator, the user selects the appropriate course of action. This selection process can be guided by suggestions based on prompts provided by a generative AI model. The user then reviews the output information on the appropriate course of action and applies it in the actual field.

[0224] 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.

[0225] 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.

[0226] 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.

[0227] [Second Embodiment]

[0228] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0229] 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.

[0230] 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).

[0231] 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.

[0232] 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.

[0233] 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).

[0234] 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.

[0235] 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.

[0236] 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.

[0237] 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.

[0238] 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.

[0239] 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".

[0240] This invention relates to a system that uses devices such as cameras and microphones to analyze human movements and voices in real time within a specific space, such as a workplace or meeting room, and detects potentially harassing behavior. This system utilizes both video and audio data to identify abnormal behavior, generate warning reports, and aims to maintain workplace safety and well-being.

[0241] Video and audio data analysis

[0242] The server receives video data from the camera and performs video analysis using machine learning. This analysis captures the movements and facial expressions of people in the video in real time and detects abnormal behavior based on specific patterns. The server also acquires audio data from microphones in the conference room and converts it into text using speech recognition software. Natural language processing techniques are then applied to this text data to identify aggressive remarks and negative emotional expressions.

[0243] Detection and reporting of abnormal behavior

[0244] When an anomaly is detected, the server immediately generates an alert report. This report contains detailed information about the abnormal behavior and is sent to the designated administrator or person in charge. For example, if a particular employee uses inappropriate language towards other participants during a meeting, the system can detect this and send a warning message to the administrator, enabling a quick response to the problem.

[0245] Monitoring and suggesting meeting progress

[0246] During meetings, the server monitors the progress of the meeting and detects situations where the discussion has stalled or a particular topic is not progressing smoothly. In such cases, the server uses AI to generate appropriate suggestions and presents them to meeting participants via their terminals. For example, if the discussion reaches an impasse, participants may be offered a suggestion such as, "What are other opinions on this topic?" to help the meeting proceed smoothly.

[0247] Feedback and evaluation

[0248] Furthermore, the system's server records and evaluates each employee's contributions and job performance during meetings. Based on this evaluation, a feedback report is generated and delivered to individual employees via their terminals. This report specifically details areas for improvement and commendable aspects, allowing each employee to use it for self-improvement.

[0249] With the configuration described above, the present invention simultaneously achieves the optimization of the workplace environment, improved employee productivity, and the cultivation of a healthy organizational culture.

[0250] The following describes the processing flow.

[0251] Step 1:

[0252] The server receives video data in real time from cameras installed in the workplace. This prepares it for continuous monitoring of people and situations within the target area.

[0253] Step 2:

[0254] The server activates a machine learning module to analyze the received video data, examining the movements and facial expressions of individuals frame by frame. This allows it to identify specific behavioral patterns and assess the likelihood of abnormal behavior.

[0255] Step 3:

[0256] The server acquires audio data from microphones in the conference room. The acquired audio data is converted into a digital format and then converted into text data by speech recognition software.

[0257] Step 4:

[0258] The server applies natural language processing algorithms to the converted text data to analyze the tone and content of the speech. This identifies offensive words and negative expressions.

[0259] Step 5:

[0260] If abnormal behavior or inappropriate remarks are detected, the server automatically generates a warning report based on that information. This report is immediately notified to administrators and responsible personnel.

[0261] Step 6:

[0262] During the meeting, the server monitors the progress and determines if the discussion is stalled or if information is lacking. If necessary, the AI ​​generates suggestions and presents solutions for improvement.

[0263] Step 7:

[0264] The generated suggestions are displayed to meeting participants via their devices, helping to ensure the meeting runs more smoothly.

[0265] Step 8:

[0266] After the meeting, the server evaluates each employee's contribution and generates a feedback report based on that data. The report is distributed to each employee via their terminal, which helps them with individual analysis and improvement.

[0267] Through this series of processes, the system helps optimize the workplace environment, prevent harassment, improve meeting efficiency, and promote employee growth.

[0268] (Example 1)

[0269] 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."

[0270] Detecting inappropriate behavior and remarks in the workplace and meetings, and responding promptly to them, has long been a challenge. Furthermore, there is a need for a system that efficiently monitors meeting progress, makes appropriate suggestions, evaluates employee contributions, and provides feedback. Improving these aspects is essential to enhancing workplace safety and productivity, and fostering a healthy organizational culture.

[0271] 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.

[0272] In this invention, the server includes means for receiving and analyzing video information, means for acquiring audio information and converting it into text information, and means for detecting abnormal behavior and generating warning reports based on the analyzed information. This makes it possible to detect inappropriate behavior in real time in the workplace or meetings and respond quickly. Furthermore, by providing means for monitoring discussions and making suggestions using AI generation, and means for evaluating job performance and generating feedback reports, it appropriately supports the efficiency of meetings and the individual performance evaluation of employees.

[0273] "Visual information" refers to visual data collected by cameras and other recording devices.

[0274] "Audio information" refers to sound data collected using microphones or other recording devices.

[0275] "Analysis" is the process of analyzing information and extracting meaningful patterns and features contained within it.

[0276] "Abnormal behavior" refers to actions or words that are unusual or considered particularly problematic.

[0277] A "warning report" is a report generated when abnormal behavior is detected, containing details of the situation and information about the need for action.

[0278] "Progress of discussion" refers to the current state of a meeting or discussion, indicating its progress and the degree of development of its content.

[0279] "Generative AI" refers to artificial intelligence systems that automatically generate data and use that data to make appropriate suggestions and decisions.

[0280] A "feedback report" is a document or information that includes an evaluation of a specific activity or outcome, as well as suggestions for improvement.

[0281] The system of the present invention performs real-time detection of harassment and abnormal behavior in the workplace environment and meeting rooms, supports the progress of meetings, and evaluates and provides feedback to employees based on video information and audio information. This system mainly has the server playing a central role and operates by combining various devices and technologies.

[0282] The server continuously collects video information and audio information through cameras and microphones. Thereby, the server monitors actions and speech in real time. For the analysis of video information, high-resolution cameras and machine learning software are used, and actions are recognized using, for example, OpenCV or TensorFlow. Through face recognition and action recognition, the actions and expressions of people are identified.

[0283] Regarding audio information, it is converted into text using speech recognition technology on the server, and then natural language processing is performed. Standard speech recognition software and natural language processing libraries such as spaCy, for example, are used in this process to identify aggressive speech and negative emotional expressions. When abnormal behavior is detected in this way, an attention report is generated on the spot and notified to the administrator or person in charge.

[0284] Furthermore, the server monitors the progress of meetings using an AI model and generates appropriate proposals when the discussion stalls. This includes a function of using a generation AI model to provide participants with proposals to facilitate conversations such as "What are other opinions on this topic?" It is expected that this will enable the meeting to proceed smoothly and draw out more opinions.

[0285] Also, the server records the contributions and work performance of each employee in meetings and conducts evaluations based on this. Based on the evaluation results, a feedback report is generated and distributed individually to employees through terminals. This report describes specific points for improvement and points of praise, which users can utilize for self-improvement.

[0286] As described above, this invention is a system that improves the quality of communication in the workplace and meetings, supports the prevention of harassment, and enhances the efficiency of performance evaluations.

[0287] The flow of the specific process in Example 1 will be described using FIG. 11.

[0288] Step 1:

[0289] The server continuously acquires video information and audio information from the camera and microphone. The input is the video and audio at the scene, which is received in a stream format. Specifically, video data from a high-resolution camera and audio data from a high-sensitivity microphone are collected. The output is this video and audio data.

[0290] Step 2:

[0291] The server analyzes the acquired video information. The input is the video data acquired in Step 1. Using a machine learning model, the actions and expressions of people in each frame are identified. Specifically, image processing is performed using libraries to perform face recognition and gesture recognition. The output is the identified people and the patterns of their actions.

[0292] Step 3:

[0293] The server converts the audio information into text. The input is the audio data acquired in Step 1. Using speech recognition technology, the audio is transcribed and prepared for application to natural language processing technology. Standard speech recognition software is used for this process. The output is the text data corresponding to the audio data.

[0294] Step 4:

[0295] The server detects aggressive remarks from the analyzed speech-to-text. The input is the text data obtained in step 3. Natural language processing techniques are used to analyze the tone and content of the words and identify abnormal remarks. Specific sentiment analysis methods are used in this process. The output is the identified aggressive words and their context.

[0296] Step 5:

[0297] The server generates a warning report based on abnormal behavior or statements and notifies the administrator. The input is the abnormal behavior or statements detected in steps 2 and 4. The warning report, including a detailed description of the situation, is sent to the relevant administrator via email or the notification system. The output is the generated warning report.

[0298] Step 6:

[0299] The server monitors the progress of the meeting and uses a generative AI model to create appropriate suggestions. The input is meeting progress data. If the discussion stalls, the AI ​​is used to generate suggestions such as "offer an opinion from a different perspective" and provides them to participants via their terminals. The output is suggestions for facilitating the smooth progress of the meeting.

[0300] Step 7:

[0301] The server evaluates employee contributions and performance and generates individual feedback reports. Input data consists of statements and actions during meetings. Based on this data, the server performs an evaluation and creates a detailed report outlining areas for improvement and areas for praise for each employee, which is then sent to each employee individually via their terminal. The output is a detailed feedback report.

[0302] (Application Example 1)

[0303] 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."

[0304] The detection of harassment and abnormal behavior in the workplace often depends on manual work or human monitoring, making it difficult to respond efficiently. In addition, the decline in productivity due to business stagnation and communication breakdown is also a serious problem. Conventional methods cannot provide quick and effective solutions to these problems. Therefore, there is a need to maintain a safe and healthy workplace environment and improve the quality of work.

[0305] The specific processing by the specific processing unit 290 of the data processing apparatus 12 in Application Example 1 is realized by the following means.

[0306] In this invention, the server includes means for receiving video information, analyzing it to identify human actions, means for acquiring audio information and converting it into text information, and means for detecting abnormal behavior based on the analyzed video information and audio information. This enables real-time detection of abnormal behavior in the workplace, prompt notification of warning reports, and visualization of abnormal behavior using an extended reality display device.

[0307] "Video information" refers to visual data acquired by devices such as cameras and sensors.

[0308] "Audio information" refers to acoustic data collected by devices such as microphones.

[0309] "Text information" refers to information expressed in characters based on audio information.

[0310] "Analysis" refers to the process of analyzing acquired data according to a specific purpose.

[0311] "Abnormal behavior" refers to behavior that deviates from normal behavior patterns and is considered problematic based on specific criteria.

[0312] "Warning report" refers to the content of the notification generated when abnormal behavior is detected and includes information to prompt appropriate responses.

[0313] An "augmented reality display device" refers to a device equipped with technology that overlays digital information onto the real world's field of view.

[0314] The system for implementing this invention consists of software and hardware that enables both video and audio processing. The server receives video and audio information from sensor devices such as cameras and microphones. The video information is analyzed in real time using video analysis libraries such as OpenCV to identify human movements. The audio information is converted into text using speech recognition software such as DeepSpeech, and then analyzed using natural language processing techniques such as NLTK. This process enables the rapid detection of abnormal behavior.

[0315] If abnormal behavior is detected, the server generates a warning report and sends it to an augmented reality display device for notification. This allows users present to visually receive information about the abnormal behavior. For example, if an employee speaks in a harsh tone at work, the system detects the audio information and displays a warning message such as "Strong language is being used" on the augmented reality display device to alert the user.

[0316] As a concrete example, focusing on the theme of "real-time detection and visualization of abnormal behavior in workplace meetings," the system analyzes video and audio to monitor the attitudes and statements of meeting participants. Using this system, users can quickly respond to subtle nuances that were previously difficult to detect. When utilizing a generative AI model, an appropriate prompt would be, "Design an optimal algorithm to analyze real-time video and audio data from the workplace and report abnormal behavior and incidents constituting harassment."

[0317] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0318] Step 1:

[0319] The server receives video and audio information acquired from the terminal via the camera and microphone. Since this input data is processed in real time, it is necessary to maintain a smooth data flow.

[0320] Step 2:

[0321] The server uses the OpenCV library to analyze video information. The purpose of the analysis is to identify human movements and facial expressions, and to extract data to identify specific abnormal behavioral patterns. The input is video information, and the output is the analyzed behavioral patterns.

[0322] Step 3:

[0323] The server uses DeepSpeech to convert audio information into text information. This conversion process extracts highly accurate text data from the audio, and sentiment analysis is performed based on this data. The input is audio information, and the output is text information.

[0324] Step 4:

[0325] The server uses NLTK to perform natural language processing on text information. Specifically, it detects aggressive remarks and negative emotional expressions in the text and flags them as abnormal behavior. The input is the transformed text information, and the output is the analysis results and the abnormal behavior flag.

[0326] Step 5:

[0327] If abnormal behavior is detected, the server generates a warning report. This report includes specific details of the detected abnormal behavior and related information. The generated report is sent to an augmented reality display device for visualization. The input is the abnormal behavior flag, and the output is the warning report.

[0328] Step 6:

[0329] The user receives a visually represented warning report through an augmented reality display device. The specific details of abnormal behavior are provided visually, serving as a tool to prompt immediate action. The input is the warning report, and the output is the user's recognition and response.

[0330] 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.

[0331] This invention is a system that detects abnormal behavior and takes appropriate action by analyzing the actions and emotions of individuals in the workplace in real time. In addition to analyzing video and audio data, this system combines an emotion engine to identify the user's emotions, contributing to improved workplace safety and employee well-being.

[0332] Video and audio data analysis

[0333] The server acquires video data from cameras placed in the workplace area and analyzes the behavior using machine learning algorithms. This makes it possible to monitor the behavior of specific individuals in real time and detect signs of abnormal behavior. The server also collects audio data from microphones and converts it into text data. Based on this text data, natural language processing technology is used to identify linguistic anomalies.

[0334] Emotion recognition by an emotion engine

[0335] The server uses an emotion engine to analyze the user's emotional state from video and audio data. The emotion engine detects facial expressions and tone of voice to determine emotions such as joy, anger, and sadness. This enables more accurate detection of abnormal behavior by incorporating emotional data.

[0336] Detection and response to abnormal behavior

[0337] If abnormal behavior or inappropriate emotional states are detected, the server immediately generates an alert report and notifies the relevant administrator or person in charge. This report records details of the behavior and emotions, helping to quickly determine appropriate action. For example, if it is detected that a participant is emotionally agitated and showing discomfort during a meeting, the situation can be reported to the administrator, who can then be instructed to take appropriate action to prevent the problem from escalating.

[0338] Post-meeting evaluation and feedback

[0339] After the meeting ends, the server analyzes data including each participant's behavior and emotions to create individual feedback reports. These reports, delivered to employees via their devices, include trends and areas for behavioral improvement based on their emotional state, promoting self-improvement.

[0340] In this way, by utilizing information about emotional states, the system can simultaneously improve employees' emotional well-being and job performance.

[0341] The following describes the processing flow.

[0342] Step 1:

[0343] The server receives video data in real time from cameras installed in the workplace and sends the data to a sequential processing stream. This prepares the data for real-time analysis.

[0344] Step 2:

[0345] The server applies machine learning models to the received video data to analyze the movements and facial expressions of individual people. It uses face recognition and motion detection algorithms to extract deviations from normal behavioral patterns.

[0346] Step 3:

[0347] The server acquires audio data from microphones installed in the conference room or designated area and stores it as digital audio. The audio processing module then passes this to the speech recognition engine, which converts it into text data.

[0348] Step 4:

[0349] The server analyzes the converted text data using natural language processing algorithms, identifying aggressive or suspicious behavior by analyzing the content and tone of the statements.

[0350] Step 5:

[0351] The server uses an emotion engine to identify emotional states from video and audio data. By analyzing changes in facial features and voice pitch, it classifies the user's emotions into categories such as "joy," "anger," and "sadness."

[0352] Step 6:

[0353] If abnormal behavior or negative emotional states are detected, the server generates an alert report summarizing the situation. This report includes detailed observations and information necessary for action, and is immediately sent to the administrator.

[0354] Step 7:

[0355] The terminal displays suggestions generated during the meeting to participants. These suggestions include specific measures to stimulate discussion based on sentiment data, thereby improving the quality of the meeting.

[0356] Step 8:

[0357] After the meeting, the server analyzes each participant's behavioral and emotional data to perform individual evaluations and generate feedback reports. These reports are distributed to employees via their devices to support their individual growth.

[0358] Step 9:

[0359] Based on the feedback received, users can develop action plans for self-improvement. The feedback serves as a guide for specific behavioral modifications.

[0360] Through these steps, the system comprehensively monitors the work environment, enabling early detection of anomalies and simultaneous improvement of productivity.

[0361] (Example 2)

[0362] 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".

[0363] In the workplace, effectively detecting and promptly responding to abnormal employee behavior and emotional changes is crucial for improving safety and operational efficiency. However, conventional systems are limited to detection based on behavioral and verbal abnormalities, making it difficult to achieve highly accurate detection and immediate response that takes emotional states into account.

[0364] 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.

[0365] In this invention, the server includes means for receiving and analyzing video data to identify a person's movements, means for acquiring audio data and converting it into text data, and means for identifying emotional states from the analyzed data. This enables highly accurate detection of emotional changes and abnormal emotions, and allows for rapid warning reporting and feedback.

[0366] "Video data" refers to visual information acquired from cameras in the workplace environment, and serves as basic data for analyzing people's movements and facial expressions.

[0367] "Means for identifying movements" refers to technologies that analyze video data to identify the movements and behavioral patterns of individual people and detect abnormal movements.

[0368] "Audio data" refers to acoustic signals acquired through a microphone, and is used to analyze conversation content and voice tone.

[0369] "Methods for converting to text data" refers to methods for converting audio data into text data, making the content analyzable.

[0370] A "means for detecting abnormal behavior" refers to a system that identifies unusual behavior or emotional changes based on video and audio data.

[0371] "Means for generating alert reports" refers to a process for documenting details of abnormal behavior or emotions when they are detected and notifying relevant parties.

[0372] "Means for identifying emotional states" refers to technologies that analyze video and audio data to determine a person's emotions (e.g., joy, anger, sadness).

[0373] "Methods for providing feedback" refer to techniques used to show participants areas for improvement and emotional tendencies based on analyzed behavioral and emotional data.

[0374] This invention is a system that analyzes human behavior and emotions in real time and detects anomalies in order to improve safety and employee well-being in the workplace environment. The main components of the invention consist of a camera, microphone, server, and terminal.

[0375] The server processes video data collected by the camera and uses TensorFlow to analyze behavior using machine learning algorithms, including open-source libraries. Through this analysis, the server identifies the behavioral state of specific individuals and detects abnormal behavior. Additionally, audio data acquired by the microphone is converted into text using speech recognition services such as Google Cloud Speech-to-Text, and natural language processing techniques are used to identify linguistic anomalies.

[0376] Furthermore, the server uses an emotion recognition API to analyze emotional states from video and audio. This analysis can detect facial expressions and voice tone, and identify emotions such as joy or anger. The emotion data improves the accuracy of detecting abnormal behavior.

[0377] When an anomaly is detected, an alert report is immediately generated and notified to the appropriate administrator. This report records specific details of the behavior and emotions, facilitating a swift response. For example, if a participant becomes emotionally unstable during a meeting, the system can detect this change and notify the administrator, preventing the problem from escalating.

[0378] Furthermore, after the meeting, the server comprehensively analyzes the participants' behavior and emotions and generates a feedback report. This report includes patterns of emotional changes and points for improving behavior, and is provided to employees via their terminals. This allows users to understand their own emotional tendencies and aim to improve their work performance.

[0379] Examples of prompt messages for this system include the following:

[0380] "Please tell me how to analyze emotional data in the workplace to detect abnormal behavior and suggest improvements."

[0381] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0382] Step 1:

[0383] The server acquires video and audio data in real time from cameras and microphones placed in the workplace. It receives video and audio data streams as input and prepares them as datasets for processing. Specific operations include temporal slicing and buffering of the data.

[0384] Step 2:

[0385] The server uses the acquired video data to analyze human movement by applying machine learning models such as TensorFlow. It uses video frames as input and decodes the action in each frame using a motion recognition algorithm. This process generates action labels as output, enabling the detection of anomalies in the movement of people and objects.

[0386] Step 3:

[0387] The server converts audio data into text data using a speech recognition service such as Google Cloud Speech-to-Text. It takes the audio signal as input, extracts acoustic features, and sequentially converts the audio into text information. The output of this process is conversation data in text format.

[0388] Step 4:

[0389] The server analyzes the converted text data using natural language processing techniques to identify linguistic anomalies. The input is text data, and semantic analysis is performed via a natural language processing engine to detect abnormal language patterns. The output includes alert information for unusual language usage.

[0390] Step 5:

[0391] The server uses an emotion recognition API to perform emotion analysis based on the analyzed video and audio. It takes image features and audio spectra as input and uses a deep learning model to classify emotion labels. The output is generated as a dataset of emotional states, identifying emotions such as joy, anger, and sadness.

[0392] Step 6:

[0393] When abnormal behavior or emotional changes are detected, the server generates an alert report and notifies administrators and other relevant parties who require attention. At this stage, the anomaly detection data is used as input, and a detailed anomaly report is created using a template for structuring the report. The generated output is sent via email or message notification.

[0394] Step 7:

[0395] After the event ends, the server analyzes data for each participant and generates feedback reports. Using behavioral and emotional datasets as input, statistical models and evaluation algorithms are employed to generate reports on emotional tendencies and areas for improvement for each employee. The final output is the report delivered via the terminal.

[0396] This processing flow enables the system to achieve real-time safety in the workplace and improve employee well-being.

[0397] (Application Example 2)

[0398] 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."

[0399] In the workplace, it is crucial to manage worker safety and mental health simultaneously and to detect abnormal behavior and emotional changes early. However, conventional systems lack the means to comprehensively analyze and notify this information in real time, resulting in a challenge in adequately addressing the improvement of workplace safety and worker well-being.

[0400] 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.

[0401] In this invention, the server includes means for receiving and analyzing video data to identify the actions of an individual, means for acquiring audio data and converting it into text data, means for detecting abnormal behavior based on the analyzed video and audio data, means for monitoring the individual's biological information in real time and analyzing their emotional state, and means for sending an alarm to a management device and instructing appropriate action based on the emotional analysis. This enables real-time monitoring of changes in the actions and emotions of workers in the workplace environment, allowing for a rapid response.

[0402] "Video data" refers to digital or analog data representing visual information acquired by a camera or other imaging device.

[0403] "Analysis" is a method of processing collected data to extract specific information or understand its characteristics.

[0404] An "individual" is an object being monitored within a system, usually referring to a human worker, but may also include other living organisms or objects.

[0405] "Action" refers to the physical actions or movements performed by an individual, and usually means bodily movements.

[0406] "Audio data" refers to sound information acquired through devices such as microphones, expressed in digital or analog format.

[0407] "Converting to text data" is the process of analyzing information such as audio and video and converting it into digital format as text.

[0408] "Detection of abnormal behavior" refers to identifying behaviors that deviate from normal behavioral patterns and recognizing the signs of such behavior.

[0409] "Biometric information" refers to data about an individual's physical state and function, including heart rate and respiratory rate.

[0410] "Analysis of emotional state" is a process that aims to evaluate an individual's psychological state and identify specific emotions such as joy or anger.

[0411] "Sending an alarm" is a systemic action to notify relevant parties of information when an anomaly or specific conditions are met.

[0412] "Management device" refers to hardware or software that controls the entire monitoring system and aggregates or distributes necessary information.

[0413] To implement this invention, a system is required in which multiple devices and software components work in cooperation. The server simultaneously acquires video and audio data from cameras and microphones in the workplace environment and analyzes them to identify the actions and emotional states of individuals. Specifically, the video data is processed in real time using an image processing library such as OpenCV, and an emotion recognition engine called EmotionEngine determines emotions from the individual's facial expressions. The audio data is converted into text data using an audio processing library such as librosa.

[0414] If the analysis detects abnormal behavior or emotional abnormalities such as stress, the server sends an alert to the management device, which then notifies the administrator's smartphone or computer. This management device uses a module called AlertManager to generate the alert. Appropriate countermeasures are automatically suggested, enabling a rapid response to the situation.

[0415] As a concrete example, when a new production line starts operation, the system automatically monitors whether employees are experiencing stress due to the change in environment. If an anomaly is detected, managers can use that information to readjust the work environment settings and implement measures to reduce the burden on workers.

[0416] To realize such a system, it is possible to obtain more detailed analysis results by supplying advanced prompt sentences to the generating AI model. For example, a prompt sentence such as, "Please give me ideas for an effective monitoring system to analyze abnormal behavior and emotions in the work environment," could be considered.

[0417] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0418] Step 1:

[0419] The server acquires video data from cameras installed in the workplace environment. It analyzes the received video data using OpenCV to identify individual movements. Specifically, it extracts the coordinates and movement features of individuals and passes this data to subsequent processing steps.

[0420] Step 2:

[0421] The server acquires audio data from microphones in the environment. This audio data is converted into text data using the librosa library. The audio waveform, as input, is analyzed into frequency components, and the content is output in text format through a speech recognition algorithm. This text data is used to detect abnormal behavior.

[0422] Step 3:

[0423] The device uses EmotionEngine to recognize emotional states from analyzed video and audio data. It analyzes facial expressions and tone of voice as input and outputs emotions such as anger and joy. This allows for the identification of signs of abnormal behavior based on emotions.

[0424] Step 4:

[0425] The server detects abnormal behavior based on analyzed behavioral and emotional data. If an abnormality is detected, it sends an alert to the management device using AlertManager. By analyzing the input abnormality information and immediately generating an alert and notifying the administrator's terminal, it encourages a quick response.

[0426] Step 5:

[0427] After receiving an alert as an administrator, the user selects the appropriate course of action. This selection process can be guided by suggestions based on prompts provided by a generative AI model. The user then reviews the output information on the appropriate course of action and applies it in the actual field.

[0428] 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.

[0429] 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.

[0430] 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.

[0431] [Third Embodiment]

[0432] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0433] 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.

[0434] 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).

[0435] 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.

[0436] 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.

[0437] 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).

[0438] 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.

[0439] 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.

[0440] 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.

[0441] 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.

[0442] 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.

[0443] 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".

[0444] This invention relates to a system that uses devices such as cameras and microphones to analyze human movements and voices in real time within a specific space, such as a workplace or meeting room, and detects potentially harassing behavior. This system utilizes both video and audio data to identify abnormal behavior, generate warning reports, and aims to maintain workplace safety and well-being.

[0445] Video and audio data analysis

[0446] The server receives video data from the camera and performs video analysis using machine learning. This analysis captures the movements and facial expressions of people in the video in real time and detects abnormal behavior based on specific patterns. The server also acquires audio data from microphones in the conference room and converts it into text using speech recognition software. Natural language processing techniques are then applied to this text data to identify aggressive remarks and negative emotional expressions.

[0447] Detection and reporting of abnormal behavior

[0448] When an anomaly is detected, the server immediately generates an alert report. This report contains detailed information about the abnormal behavior and is sent to the designated administrator or person in charge. For example, if a particular employee uses inappropriate language towards other participants during a meeting, the system can detect this and send a warning message to the administrator, enabling a quick response to the problem.

[0449] Monitoring and suggesting meeting progress

[0450] During meetings, the server monitors the progress of the meeting and detects situations where the discussion has stalled or a particular topic is not progressing smoothly. In such cases, the server uses AI to generate appropriate suggestions and presents them to meeting participants via their terminals. For example, if the discussion reaches an impasse, participants may be offered a suggestion such as, "What are other opinions on this topic?" to help the meeting proceed smoothly.

[0451] Feedback and evaluation

[0452] Furthermore, the system's server records and evaluates each employee's contributions and job performance during meetings. Based on this evaluation, a feedback report is generated and delivered to individual employees via their terminals. This report specifically details areas for improvement and commendable aspects, allowing each employee to use it for self-improvement.

[0453] With the configuration described above, the present invention simultaneously achieves the optimization of the workplace environment, improved employee productivity, and the cultivation of a healthy organizational culture.

[0454] The following describes the processing flow.

[0455] Step 1:

[0456] The server receives video data in real time from cameras installed in the workplace. This prepares it for continuous monitoring of people and situations within the target area.

[0457] Step 2:

[0458] The server activates a machine learning module to analyze the received video data, examining the movements and facial expressions of individuals frame by frame. This allows it to identify specific behavioral patterns and assess the likelihood of abnormal behavior.

[0459] Step 3:

[0460] The server acquires audio data from microphones in the conference room. The acquired audio data is converted into a digital format and then converted into text data by speech recognition software.

[0461] Step 4:

[0462] The server applies natural language processing algorithms to the converted text data to analyze the tone and content of the speech. This identifies offensive words and negative expressions.

[0463] Step 5:

[0464] If abnormal behavior or inappropriate remarks are detected, the server automatically generates a warning report based on that information. This report is immediately notified to administrators and responsible personnel.

[0465] Step 6:

[0466] During the meeting, the server monitors the progress and determines if the discussion is stalled or if information is lacking. If necessary, the AI ​​generates suggestions and presents solutions for improvement.

[0467] Step 7:

[0468] The generated suggestions are displayed to meeting participants via their devices, helping to ensure the meeting runs more smoothly.

[0469] Step 8:

[0470] After the meeting, the server evaluates each employee's contribution and generates a feedback report based on that data. The report is distributed to each employee via their terminal, which helps them with individual analysis and improvement.

[0471] Through this series of processes, the system helps optimize the workplace environment, prevent harassment, improve meeting efficiency, and promote employee growth.

[0472] (Example 1)

[0473] 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."

[0474] Detecting inappropriate behavior and remarks in the workplace and meetings, and responding promptly to them, has long been a challenge. Furthermore, there is a need for a system that efficiently monitors meeting progress, makes appropriate suggestions, evaluates employee contributions, and provides feedback. Improving these aspects is essential to enhancing workplace safety and productivity, and fostering a healthy organizational culture.

[0475] 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.

[0476] In this invention, the server includes means for receiving and analyzing video information, means for acquiring audio information and converting it into text information, and means for detecting abnormal behavior and generating warning reports based on the analyzed information. This makes it possible to detect inappropriate behavior in real time in the workplace or meetings and respond quickly. Furthermore, by providing means for monitoring discussions and making suggestions using AI generation, and means for evaluating job performance and generating feedback reports, it appropriately supports the efficiency of meetings and the individual performance evaluation of employees.

[0477] "Visual information" refers to visual data collected by cameras and other recording devices.

[0478] "Audio information" refers to sound data collected using microphones or other recording devices.

[0479] "Analysis" is the process of analyzing information and extracting meaningful patterns and features contained within it.

[0480] "Abnormal behavior" refers to actions or words that are unusual or considered particularly problematic.

[0481] A "warning report" is a report generated when abnormal behavior is detected, containing details of the situation and information about the need for action.

[0482] "Progress of discussion" refers to the current state of a meeting or discussion, indicating its progress and the degree of development of its content.

[0483] "Generative AI" refers to artificial intelligence systems that automatically generate data and use that data to make appropriate suggestions and decisions.

[0484] A "feedback report" is a document or information that includes an evaluation of a specific activity or outcome, as well as suggestions for improvement.

[0485] The present invention provides real-time detection of harassment and abnormal behavior in the workplace and meeting rooms, support for meeting progress, and employee evaluation and feedback based on video and audio information. This system primarily relies on a server to operate, combining various devices and technologies.

[0486] The server continuously collects video and audio information through cameras and microphones. This allows the server to monitor actions and speech in real time. High-resolution cameras and machine learning software are used to analyze the video information, for example, using OpenCV or TensorFlow to recognize actions. Face recognition and behavior recognition identify a person's actions and facial expressions.

[0487] Audio information is converted to text using speech recognition technology on the server, and then natural language processing is performed. This process uses standard speech recognition software or natural language processing libraries such as spaCy to identify aggressive remarks and negative emotional expressions. As a result, if abnormal behavior is detected, a warning report is generated immediately and notified to administrators or responsible personnel.

[0488] Furthermore, the server monitors the meeting's progress using an AI model and generates appropriate suggestions when the discussion stalls. This includes a feature that uses a generative AI model to provide participants with suggestions to facilitate conversation, such as, "What are other opinions on this topic?" This is expected to ensure the meeting runs smoothly and elicit more opinions.

[0489] The server also records each employee's contributions to meetings and their work performance, and evaluates them based on this data. Based on these evaluations, a feedback report is generated and delivered individually to each employee via their terminal. This report includes specific areas for improvement and points of praise, which users can use to improve themselves.

[0490] As described above, this invention is a system that improves the quality of communication in the workplace and meetings, and supports the prevention of harassment and the efficiency of job evaluations.

[0491] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0492] Step 1:

[0493] The server continuously acquires video and audio information from cameras and microphones. The input is on-site video and audio, which is received in stream format. Specifically, it collects video data from high-resolution cameras and audio data from high-sensitivity microphones. The output is this video and audio data.

[0494] Step 2:

[0495] The server analyzes the acquired video information. The input is the video data acquired in step 1. A machine learning model is used to identify the actions and facial expressions of people in each frame. Specifically, image processing is performed using libraries to perform face recognition and gesture recognition. The output is the identified people and their action patterns.

[0496] Step 3:

[0497] The server converts the audio information into text. The input is the audio data obtained in step 1. Speech recognition technology is used to transcribe the audio into text, preparing it for application to natural language processing technology. Standard speech recognition software is used for this process. The output is the text data corresponding to the audio data.

[0498] Step 4:

[0499] The server detects aggressive remarks from the analyzed speech-to-text. The input is the text data obtained in step 3. Natural language processing techniques are used to analyze the tone and content of the words and identify abnormal remarks. Specific sentiment analysis methods are used in this process. The output is the identified aggressive words and their context.

[0500] Step 5:

[0501] The server generates a warning report based on abnormal behavior or statements and notifies the administrator. The input is the abnormal behavior or statements detected in steps 2 and 4. The warning report, including a detailed description of the situation, is sent to the relevant administrator via email or the notification system. The output is the generated warning report.

[0502] Step 6:

[0503] The server monitors the progress of the meeting and uses a generative AI model to create appropriate suggestions. The input is meeting progress data. If the discussion stalls, the AI ​​is used to generate suggestions such as "offer an opinion from a different perspective" and provides them to participants via their terminals. The output is suggestions for facilitating the smooth progress of the meeting.

[0504] Step 7:

[0505] The server evaluates employee contributions and performance and generates individual feedback reports. Input data consists of statements and actions during meetings. Based on this data, the server performs an evaluation and creates a detailed report outlining areas for improvement and areas for praise for each employee, which is then sent to each employee individually via their terminal. The output is a detailed feedback report.

[0506] (Application Example 1)

[0507] 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."

[0508] Detecting harassment and abnormal behavior in the workplace often relies on manual processes and human monitoring, making efficient responses difficult. Furthermore, decreased productivity due to work stagnation and communication breakdowns is a serious problem. Traditional methods cannot provide rapid and effective solutions to these issues. Therefore, there is a need to maintain a safe and healthy workplace environment and improve the quality of work.

[0509] 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.

[0510] In this invention, the server includes means for receiving and analyzing video information to identify human movements, means for acquiring audio information and converting it into text information, and means for detecting abnormal behavior based on the analyzed video and audio information. This enables real-time detection of abnormal behavior in the workplace, rapid notification of warning reports, and visualization of abnormal behavior using an augmented reality display device.

[0511] "Visual information" refers to visual data acquired by devices such as cameras and sensors.

[0512] "Audio information" refers to acoustic data collected by devices such as microphones.

[0513] "Text information" refers to information that is expressed in written form based on audio information.

[0514] "Analysis" refers to the process of using acquired data to analyze it according to a specific purpose.

[0515] "Abnormal behavior" refers to behavior that deviates from normal behavioral patterns and is considered problematic based on specific criteria.

[0516] A "warning report" refers to the notification content generated when abnormal behavior is detected, and includes information to prompt appropriate action.

[0517] An "augmented reality display device" refers to a device equipped with technology that overlays digital information onto the real world's field of view.

[0518] The system for implementing this invention consists of software and hardware that enables both video and audio processing. The server receives video and audio information from sensor devices such as cameras and microphones. The video information is analyzed in real time using video analysis libraries such as OpenCV to identify human movements. The audio information is converted into text using speech recognition software such as DeepSpeech, and then analyzed using natural language processing techniques such as NLTK. This process enables the rapid detection of abnormal behavior.

[0519] If abnormal behavior is detected, the server generates a warning report and sends it to an augmented reality display device for notification. This allows users present to visually receive information about the abnormal behavior. For example, if an employee speaks in a harsh tone at work, the system detects the audio information and displays a warning message such as "Strong language is being used" on the augmented reality display device to alert the user.

[0520] As a concrete example, focusing on the theme of "real-time detection and visualization of abnormal behavior in workplace meetings," the system analyzes video and audio to monitor the attitudes and statements of meeting participants. Using this system, users can quickly respond to subtle nuances that were previously difficult to detect. When utilizing a generative AI model, an appropriate prompt would be, "Design an optimal algorithm to analyze real-time video and audio data from the workplace and report abnormal behavior and incidents constituting harassment."

[0521] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0522] Step 1:

[0523] The server receives video and audio information acquired from the terminal via the camera and microphone. Since this input data is processed in real time, it is necessary to maintain a smooth data flow.

[0524] Step 2:

[0525] The server uses the OpenCV library to analyze video information. The purpose of the analysis is to identify human movements and facial expressions, and to extract data to identify specific abnormal behavioral patterns. The input is video information, and the output is the analyzed behavioral patterns.

[0526] Step 3:

[0527] The server uses DeepSpeech to convert audio information into text information. This conversion process extracts highly accurate text data from the audio, and sentiment analysis is performed based on this data. The input is audio information, and the output is text information.

[0528] Step 4:

[0529] The server uses NLTK to perform natural language processing on text information. Specifically, it detects aggressive remarks and negative emotional expressions in the text and flags them as abnormal behavior. The input is the transformed text information, and the output is the analysis results and the abnormal behavior flag.

[0530] Step 5:

[0531] If abnormal behavior is detected, the server generates a warning report. This report includes specific details of the detected abnormal behavior and related information. The generated report is sent to an augmented reality display device for visualization. The input is the abnormal behavior flag, and the output is the warning report.

[0532] Step 6:

[0533] The user receives a visually represented warning report through an augmented reality display device. The specific details of abnormal behavior are provided visually, serving as a tool to prompt immediate action. The input is the warning report, and the output is the user's recognition and response.

[0534] 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.

[0535] This invention is a system that detects abnormal behavior and takes appropriate action by analyzing the actions and emotions of individuals in the workplace in real time. In addition to analyzing video and audio data, this system combines an emotion engine to identify the user's emotions, contributing to improved workplace safety and employee well-being.

[0536] Video and audio data analysis

[0537] The server acquires video data from cameras placed in the workplace area and analyzes the behavior using machine learning algorithms. This makes it possible to monitor the behavior of specific individuals in real time and detect signs of abnormal behavior. The server also collects audio data from microphones and converts it into text data. Based on this text data, natural language processing technology is used to identify linguistic anomalies.

[0538] Emotion recognition by an emotion engine

[0539] The server uses an emotion engine to analyze the user's emotional state from video and audio data. The emotion engine detects facial expressions and tone of voice to determine emotions such as joy, anger, and sadness. This enables more accurate detection of abnormal behavior by incorporating emotional data.

[0540] Detection and response to abnormal behavior

[0541] If abnormal behavior or inappropriate emotional states are detected, the server immediately generates an alert report and notifies the relevant administrator or person in charge. This report records details of the behavior and emotions, helping to quickly determine appropriate action. For example, if it is detected that a participant is emotionally agitated and showing discomfort during a meeting, the situation can be reported to the administrator, who can then be instructed to take appropriate action to prevent the problem from escalating.

[0542] Post-meeting evaluation and feedback

[0543] After the meeting ends, the server analyzes data including each participant's behavior and emotions to create individual feedback reports. These reports, delivered to employees via their devices, include trends and areas for behavioral improvement based on their emotional state, promoting self-improvement.

[0544] In this way, by utilizing information about emotional states, the system can simultaneously improve employees' emotional well-being and job performance.

[0545] The following describes the processing flow.

[0546] Step 1:

[0547] The server receives video data in real time from cameras installed in the workplace and sends the data to a sequential processing stream. This prepares the data for real-time analysis.

[0548] Step 2:

[0549] The server applies machine learning models to the received video data to analyze the movements and facial expressions of individual people. It uses face recognition and motion detection algorithms to extract deviations from normal behavioral patterns.

[0550] Step 3:

[0551] The server acquires audio data from microphones installed in the conference room or designated area and stores it as digital audio. The audio processing module then passes this to the speech recognition engine, which converts it into text data.

[0552] Step 4:

[0553] The server analyzes the converted text data using natural language processing algorithms, identifying aggressive or suspicious behavior by analyzing the content and tone of the statements.

[0554] Step 5:

[0555] The server uses an emotion engine to identify emotional states from video and audio data. By analyzing changes in facial features and voice pitch, it classifies the user's emotions into categories such as "joy," "anger," and "sadness."

[0556] Step 6:

[0557] If abnormal behavior or negative emotional states are detected, the server generates an alert report summarizing the situation. This report includes detailed observations and information necessary for action, and is immediately sent to the administrator.

[0558] Step 7:

[0559] The terminal displays suggestions generated during the meeting to participants. These suggestions include specific measures to stimulate discussion based on sentiment data, thereby improving the quality of the meeting.

[0560] Step 8:

[0561] After the meeting, the server analyzes each participant's behavioral and emotional data to perform individual evaluations and generate feedback reports. These reports are distributed to employees via their devices to support their individual growth.

[0562] Step 9:

[0563] Based on the feedback received, users can develop action plans for self-improvement. The feedback serves as a guide for specific behavioral modifications.

[0564] Through these steps, the system comprehensively monitors the work environment, enabling early detection of anomalies and simultaneous improvement of productivity.

[0565] (Example 2)

[0566] 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."

[0567] In the workplace, effectively detecting and promptly responding to abnormal employee behavior and emotional changes is crucial for improving safety and operational efficiency. However, conventional systems are limited to detection based on behavioral and verbal abnormalities, making it difficult to achieve highly accurate detection and immediate response that takes emotional states into account.

[0568] 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.

[0569] In this invention, the server includes means for receiving and analyzing video data to identify a person's movements, means for acquiring audio data and converting it into text data, and means for identifying emotional states from the analyzed data. This enables highly accurate detection of emotional changes and abnormal emotions, and allows for rapid warning reporting and feedback.

[0570] "Video data" refers to visual information acquired from cameras in the workplace environment, and serves as basic data for analyzing people's movements and facial expressions.

[0571] "Means for identifying movements" refers to technologies that analyze video data to identify the movements and behavioral patterns of individual people and detect abnormal movements.

[0572] "Audio data" refers to acoustic signals acquired through a microphone, and is used to analyze conversation content and voice tone.

[0573] "Methods for converting to text data" refers to methods for converting audio data into text data, making the content analyzable.

[0574] A "means for detecting abnormal behavior" refers to a system that identifies unusual behavior or emotional changes based on video and audio data.

[0575] "Means for generating alert reports" refers to a process for documenting details of abnormal behavior or emotions when they are detected and notifying relevant parties.

[0576] "Means for identifying emotional states" refers to technologies that analyze video and audio data to determine a person's emotions (e.g., joy, anger, sadness).

[0577] "Methods for providing feedback" refer to techniques used to show participants areas for improvement and emotional tendencies based on analyzed behavioral and emotional data.

[0578] This invention is a system that analyzes human behavior and emotions in real time and detects anomalies in order to improve safety and employee well-being in the workplace environment. The main components of the invention consist of a camera, microphone, server, and terminal.

[0579] The server processes video data collected by the camera and uses TensorFlow to analyze behavior using machine learning algorithms, including open-source libraries. Through this analysis, the server identifies the behavioral state of specific individuals and detects abnormal behavior. Additionally, audio data acquired by the microphone is converted into text using speech recognition services such as Google Cloud Speech-to-Text, and natural language processing techniques are used to identify linguistic anomalies.

[0580] Furthermore, the server uses an emotion recognition API to analyze emotional states from video and audio. This analysis can detect facial expressions and voice tone, and identify emotions such as joy or anger. The emotion data improves the accuracy of detecting abnormal behavior.

[0581] When an anomaly is detected, an alert report is immediately generated and notified to the appropriate administrator. This report records specific details of the behavior and emotions, facilitating a swift response. For example, if a participant becomes emotionally unstable during a meeting, the system can detect this change and notify the administrator, preventing the problem from escalating.

[0582] Furthermore, after the meeting, the server comprehensively analyzes the participants' behavior and emotions and generates a feedback report. This report includes patterns of emotional changes and points for improving behavior, and is provided to employees via their terminals. This allows users to understand their own emotional tendencies and aim to improve their work performance.

[0583] Examples of prompt messages for this system include the following:

[0584] "Please tell me how to analyze emotional data in the workplace to detect abnormal behavior and suggest improvements."

[0585] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0586] Step 1:

[0587] The server acquires video and audio data in real time from cameras and microphones placed in the workplace. It receives video and audio data streams as input and prepares them as datasets for processing. Specific operations include temporal slicing and buffering of the data.

[0588] Step 2:

[0589] The server uses the acquired video data to analyze human movement by applying machine learning models such as TensorFlow. It uses video frames as input and decodes the action in each frame using a motion recognition algorithm. This process generates action labels as output, enabling the detection of anomalies in the movement of people and objects.

[0590] Step 3:

[0591] The server converts audio data into text data using a speech recognition service such as Google Cloud Speech-to-Text. It takes the audio signal as input, extracts acoustic features, and sequentially converts the audio into text information. The output of this process is conversation data in text format.

[0592] Step 4:

[0593] The server analyzes the converted text data using natural language processing techniques to identify linguistic anomalies. The input is text data, and semantic analysis is performed via a natural language processing engine to detect abnormal language patterns. The output includes alert information for unusual language usage.

[0594] Step 5:

[0595] The server uses an emotion recognition API to perform emotion analysis based on the analyzed video and audio. It takes image features and audio spectra as input and uses a deep learning model to classify emotion labels. The output is generated as a dataset of emotional states, identifying emotions such as joy, anger, and sadness.

[0596] Step 6:

[0597] When abnormal behavior or emotional changes are detected, the server generates an alert report and notifies administrators and other relevant parties who require attention. At this stage, the anomaly detection data is used as input, and a detailed anomaly report is created using a template for structuring the report. The generated output is sent via email or message notification.

[0598] Step 7:

[0599] After the event ends, the server analyzes data for each participant and generates feedback reports. Using behavioral and emotional datasets as input, statistical models and evaluation algorithms are employed to generate reports on emotional tendencies and areas for improvement for each employee. The final output is the report delivered via the terminal.

[0600] This processing flow enables the system to achieve real-time safety in the workplace and improve employee well-being.

[0601] (Application Example 2)

[0602] 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."

[0603] In the workplace, it is crucial to manage worker safety and mental health simultaneously and to detect abnormal behavior and emotional changes early. However, conventional systems lack the means to comprehensively analyze and notify this information in real time, resulting in a challenge in adequately addressing the improvement of workplace safety and worker well-being.

[0604] 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.

[0605] In this invention, the server includes means for receiving and analyzing video data to identify the actions of an individual, means for acquiring audio data and converting it into text data, means for detecting abnormal behavior based on the analyzed video and audio data, means for monitoring the individual's biological information in real time and analyzing their emotional state, and means for sending an alarm to a management device and instructing appropriate action based on the emotional analysis. This enables real-time monitoring of changes in the actions and emotions of workers in the workplace environment, allowing for a rapid response.

[0606] "Video data" refers to digital or analog data representing visual information acquired by a camera or other imaging device.

[0607] "Analysis" is a method of processing collected data to extract specific information or understand its characteristics.

[0608] An "individual" is an object being monitored within a system, usually referring to a human worker, but may also include other living organisms or objects.

[0609] "Action" refers to the physical actions or movements performed by an individual, and usually means bodily movements.

[0610] "Audio data" refers to sound information acquired through devices such as microphones, expressed in digital or analog format.

[0611] "Converting to text data" is the process of analyzing information such as audio and video and converting it into digital format as text.

[0612] "Detection of abnormal behavior" refers to identifying behaviors that deviate from normal behavioral patterns and recognizing the signs of such behavior.

[0613] "Biometric information" refers to data about an individual's physical state and function, including heart rate and respiratory rate.

[0614] "Analysis of emotional state" is a process that aims to evaluate an individual's psychological state and identify specific emotions such as joy or anger.

[0615] "Sending an alarm" is a systemic action to notify relevant parties of information when an anomaly or specific conditions are met.

[0616] "Management device" refers to hardware or software that controls the entire monitoring system and aggregates or distributes necessary information.

[0617] To implement this invention, a system is required in which multiple devices and software components work in cooperation. The server simultaneously acquires video and audio data from cameras and microphones in the workplace environment and analyzes them to identify the actions and emotional states of individuals. Specifically, the video data is processed in real time using an image processing library such as OpenCV, and an emotion recognition engine called EmotionEngine determines emotions from the individual's facial expressions. The audio data is converted into text data using an audio processing library such as librosa.

[0618] If the analysis detects abnormal behavior or emotional abnormalities such as stress, the server sends an alert to the management device, which then notifies the administrator's smartphone or computer. This management device uses a module called AlertManager to generate the alert. Appropriate countermeasures are automatically suggested, enabling a rapid response to the situation.

[0619] As a concrete example, when a new production line starts operation, the system automatically monitors whether employees are experiencing stress due to the change in environment. If an anomaly is detected, managers can use that information to readjust the work environment settings and implement measures to reduce the burden on workers.

[0620] To realize such a system, it is possible to obtain more detailed analysis results by supplying advanced prompt sentences to the generating AI model. For example, a prompt sentence such as, "Please give me ideas for an effective monitoring system to analyze abnormal behavior and emotions in the work environment," could be considered.

[0621] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0622] Step 1:

[0623] The server acquires video data from cameras installed in the workplace environment. It analyzes the received video data using OpenCV to identify individual movements. Specifically, it extracts the coordinates and movement features of individuals and passes this data to subsequent processing steps.

[0624] Step 2:

[0625] The server acquires audio data from microphones in the environment. This audio data is converted into text data using the librosa library. The audio waveform, as input, is analyzed into frequency components, and the content is output in text format through a speech recognition algorithm. This text data is used to detect abnormal behavior.

[0626] Step 3:

[0627] The device uses EmotionEngine to recognize emotional states from analyzed video and audio data. It analyzes facial expressions and tone of voice as input and outputs emotions such as anger and joy. This allows for the identification of signs of abnormal behavior based on emotions.

[0628] Step 4:

[0629] The server detects abnormal behavior based on analyzed behavioral and emotional data. If an abnormality is detected, it sends an alert to the management device using AlertManager. By analyzing the input abnormality information and immediately generating an alert and notifying the administrator's terminal, it encourages a quick response.

[0630] Step 5:

[0631] After receiving an alert as an administrator, the user selects the appropriate course of action. This selection process can be guided by suggestions based on prompts provided by a generative AI model. The user then reviews the output information on the appropriate course of action and applies it in the actual field.

[0632] 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.

[0633] 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.

[0634] 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.

[0635] [Fourth Embodiment]

[0636] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0637] 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.

[0638] 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).

[0639] 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.

[0640] 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.

[0641] 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).

[0642] 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.

[0643] 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.

[0644] 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.

[0645] 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.

[0646] 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.

[0647] 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.

[0648] 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".

[0649] This invention relates to a system that uses devices such as cameras and microphones to analyze human movements and voices in real time within a specific space, such as a workplace or meeting room, and detects potentially harassing behavior. This system utilizes both video and audio data to identify abnormal behavior, generate warning reports, and aims to maintain workplace safety and well-being.

[0650] Video and audio data analysis

[0651] The server receives video data from the camera and performs video analysis using machine learning. This analysis captures the movements and facial expressions of people in the video in real time and detects abnormal behavior based on specific patterns. The server also acquires audio data from microphones in the conference room and converts it into text using speech recognition software. Natural language processing techniques are then applied to this text data to identify aggressive remarks and negative emotional expressions.

[0652] Detection and reporting of abnormal behavior

[0653] When an anomaly is detected, the server immediately generates an alert report. This report contains detailed information about the abnormal behavior and is sent to the designated administrator or person in charge. For example, if a particular employee uses inappropriate language towards other participants during a meeting, the system can detect this and send a warning message to the administrator, enabling a quick response to the problem.

[0654] Monitoring and suggesting meeting progress

[0655] During meetings, the server monitors the progress of the meeting and detects situations where the discussion has stalled or a particular topic is not progressing smoothly. In such cases, the server uses AI to generate appropriate suggestions and presents them to meeting participants via their terminals. For example, if the discussion reaches an impasse, participants may be offered a suggestion such as, "What are other opinions on this topic?" to help the meeting proceed smoothly.

[0656] Feedback and evaluation

[0657] Furthermore, the system's server records and evaluates each employee's contributions and job performance during meetings. Based on this evaluation, a feedback report is generated and delivered to individual employees via their terminals. This report specifically details areas for improvement and commendable aspects, allowing each employee to use it for self-improvement.

[0658] With the configuration described above, the present invention simultaneously achieves the optimization of the workplace environment, improved employee productivity, and the cultivation of a healthy organizational culture.

[0659] The following describes the processing flow.

[0660] Step 1:

[0661] The server receives video data in real time from cameras installed in the workplace. This prepares it for continuous monitoring of people and situations within the target area.

[0662] Step 2:

[0663] The server activates a machine learning module to analyze the received video data, examining the movements and facial expressions of individuals frame by frame. This allows it to identify specific behavioral patterns and assess the likelihood of abnormal behavior.

[0664] Step 3:

[0665] The server acquires audio data from microphones in the conference room. The acquired audio data is converted into a digital format and then converted into text data by speech recognition software.

[0666] Step 4:

[0667] The server applies natural language processing algorithms to the converted text data to analyze the tone and content of the speech. This identifies offensive words and negative expressions.

[0668] Step 5:

[0669] If abnormal behavior or inappropriate remarks are detected, the server automatically generates a warning report based on that information. This report is immediately notified to administrators and responsible personnel.

[0670] Step 6:

[0671] During the meeting, the server monitors the progress and determines if the discussion is stalled or if information is lacking. If necessary, the AI ​​generates suggestions and presents solutions for improvement.

[0672] Step 7:

[0673] The generated suggestions are displayed to meeting participants via their devices, helping to ensure the meeting runs more smoothly.

[0674] Step 8:

[0675] After the meeting, the server evaluates each employee's contribution and generates a feedback report based on that data. The report is distributed to each employee via their terminal, which helps them with individual analysis and improvement.

[0676] Through this series of processes, the system helps optimize the workplace environment, prevent harassment, improve meeting efficiency, and promote employee growth.

[0677] (Example 1)

[0678] 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".

[0679] Detecting inappropriate behavior and remarks in the workplace and meetings, and responding promptly to them, has long been a challenge. Furthermore, there is a need for a system that efficiently monitors meeting progress, makes appropriate suggestions, evaluates employee contributions, and provides feedback. Improving these aspects is essential to enhancing workplace safety and productivity, and fostering a healthy organizational culture.

[0680] 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.

[0681] In this invention, the server includes means for receiving and analyzing video information, means for acquiring audio information and converting it into text information, and means for detecting abnormal behavior and generating warning reports based on the analyzed information. This makes it possible to detect inappropriate behavior in real time in the workplace or meetings and respond quickly. Furthermore, by providing means for monitoring discussions and making suggestions using AI generation, and means for evaluating job performance and generating feedback reports, it appropriately supports the efficiency of meetings and the individual performance evaluation of employees.

[0682] "Visual information" refers to visual data collected by cameras and other recording devices.

[0683] "Audio information" refers to sound data collected using microphones or other recording devices.

[0684] "Analysis" is the process of analyzing information and extracting meaningful patterns and features contained within it.

[0685] "Abnormal behavior" refers to actions or words that are unusual or considered particularly problematic.

[0686] A "warning report" is a report generated when abnormal behavior is detected, containing details of the situation and information about the need for action.

[0687] "Progress of discussion" refers to the current state of a meeting or discussion, indicating its progress and the degree of development of its content.

[0688] "Generative AI" refers to artificial intelligence systems that automatically generate data and use that data to make appropriate suggestions and decisions.

[0689] A "feedback report" is a document or information that includes an evaluation of a specific activity or outcome, as well as suggestions for improvement.

[0690] The present invention provides real-time detection of harassment and abnormal behavior in the workplace and meeting rooms, support for meeting progress, and employee evaluation and feedback based on video and audio information. This system primarily relies on a server to operate, combining various devices and technologies.

[0691] The server continuously collects video and audio information through cameras and microphones. This allows the server to monitor actions and speech in real time. High-resolution cameras and machine learning software are used to analyze the video information, for example, using OpenCV or TensorFlow to recognize actions. Face recognition and behavior recognition identify a person's actions and facial expressions.

[0692] Audio information is converted to text using speech recognition technology on the server, and then natural language processing is performed. This process uses standard speech recognition software or natural language processing libraries such as spaCy to identify aggressive remarks and negative emotional expressions. As a result, if abnormal behavior is detected, a warning report is generated immediately and notified to administrators or responsible personnel.

[0693] Furthermore, the server monitors the meeting's progress using an AI model and generates appropriate suggestions when the discussion stalls. This includes a feature that uses a generative AI model to provide participants with suggestions to facilitate conversation, such as, "What are other opinions on this topic?" This is expected to ensure the meeting runs smoothly and elicit more opinions.

[0694] The server also records each employee's contributions to meetings and their work performance, and evaluates them based on this data. Based on these evaluations, a feedback report is generated and delivered individually to each employee via their terminal. This report includes specific areas for improvement and points of praise, which users can use to improve themselves.

[0695] As described above, this invention is a system that improves the quality of communication in the workplace and meetings, and supports the prevention of harassment and the efficiency of job evaluations.

[0696] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0697] Step 1:

[0698] The server continuously acquires video and audio information from cameras and microphones. The input is on-site video and audio, which is received in stream format. Specifically, it collects video data from high-resolution cameras and audio data from high-sensitivity microphones. The output is this video and audio data.

[0699] Step 2:

[0700] The server analyzes the acquired video information. The input is the video data acquired in step 1. A machine learning model is used to identify the actions and facial expressions of people in each frame. Specifically, image processing is performed using libraries to perform face recognition and gesture recognition. The output is the identified people and their action patterns.

[0701] Step 3:

[0702] The server converts the audio information into text. The input is the audio data obtained in step 1. Speech recognition technology is used to transcribe the audio into text, preparing it for application to natural language processing technology. Standard speech recognition software is used for this process. The output is the text data corresponding to the audio data.

[0703] Step 4:

[0704] The server detects aggressive remarks from the analyzed speech-to-text. The input is the text data obtained in step 3. Natural language processing techniques are used to analyze the tone and content of the words and identify abnormal remarks. Specific sentiment analysis methods are used in this process. The output is the identified aggressive words and their context.

[0705] Step 5:

[0706] The server generates a warning report based on abnormal behavior or statements and notifies the administrator. The input is the abnormal behavior or statements detected in steps 2 and 4. The warning report, including a detailed description of the situation, is sent to the relevant administrator via email or the notification system. The output is the generated warning report.

[0707] Step 6:

[0708] The server monitors the progress of the meeting and uses a generative AI model to create appropriate suggestions. The input is meeting progress data. If the discussion stalls, the AI ​​is used to generate suggestions such as "offer an opinion from a different perspective" and provides them to participants via their terminals. The output is suggestions for facilitating the smooth progress of the meeting.

[0709] Step 7:

[0710] The server evaluates employee contributions and performance and generates individual feedback reports. Input data consists of statements and actions during meetings. Based on this data, the server performs an evaluation and creates a detailed report outlining areas for improvement and areas for praise for each employee, which is then sent to each employee individually via their terminal. The output is a detailed feedback report.

[0711] (Application Example 1)

[0712] 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".

[0713] Detecting harassment and abnormal behavior in the workplace often relies on manual processes and human monitoring, making efficient responses difficult. Furthermore, decreased productivity due to work stagnation and communication breakdowns is a serious problem. Traditional methods cannot provide rapid and effective solutions to these issues. Therefore, there is a need to maintain a safe and healthy workplace environment and improve the quality of work.

[0714] 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.

[0715] In this invention, the server includes means for receiving and analyzing video information to identify human movements, means for acquiring audio information and converting it into text information, and means for detecting abnormal behavior based on the analyzed video and audio information. This enables real-time detection of abnormal behavior in the workplace, rapid notification of warning reports, and visualization of abnormal behavior using an augmented reality display device.

[0716] "Visual information" refers to visual data acquired by devices such as cameras and sensors.

[0717] "Audio information" refers to acoustic data collected by devices such as microphones.

[0718] "Text information" refers to information that is expressed in written form based on audio information.

[0719] "Analysis" refers to the process of using acquired data to analyze it according to a specific purpose.

[0720] "Abnormal behavior" refers to behavior that deviates from normal behavioral patterns and is considered problematic based on specific criteria.

[0721] A "warning report" refers to the notification content generated when abnormal behavior is detected, and includes information to prompt appropriate action.

[0722] An "augmented reality display device" refers to a device equipped with technology that overlays digital information onto the real world's field of view.

[0723] The system for implementing this invention consists of software and hardware that enables both video and audio processing. The server receives video and audio information from sensor devices such as cameras and microphones. The video information is analyzed in real time using video analysis libraries such as OpenCV to identify human movements. The audio information is converted into text using speech recognition software such as DeepSpeech, and then analyzed using natural language processing techniques such as NLTK. This process enables the rapid detection of abnormal behavior.

[0724] If abnormal behavior is detected, the server generates a warning report and sends it to an augmented reality display device for notification. This allows users present to visually receive information about the abnormal behavior. For example, if an employee speaks in a harsh tone at work, the system detects the audio information and displays a warning message such as "Strong language is being used" on the augmented reality display device to alert the user.

[0725] As a concrete example, focusing on the theme of "real-time detection and visualization of abnormal behavior in workplace meetings," the system analyzes video and audio to monitor the attitudes and statements of meeting participants. Using this system, users can quickly respond to subtle nuances that were previously difficult to detect. When utilizing a generative AI model, an appropriate prompt would be, "Design an optimal algorithm to analyze real-time video and audio data from the workplace and report abnormal behavior and incidents constituting harassment."

[0726] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0727] Step 1:

[0728] The server receives video and audio information acquired from the terminal via the camera and microphone. Since this input data is processed in real time, it is necessary to maintain a smooth data flow.

[0729] Step 2:

[0730] The server uses the OpenCV library to analyze video information. The purpose of the analysis is to identify human movements and facial expressions, and to extract data to identify specific abnormal behavioral patterns. The input is video information, and the output is the analyzed behavioral patterns.

[0731] Step 3:

[0732] The server uses DeepSpeech to convert audio information into text information. This conversion process extracts highly accurate text data from the audio, and sentiment analysis is performed based on this data. The input is audio information, and the output is text information.

[0733] Step 4:

[0734] The server uses NLTK to perform natural language processing on text information. Specifically, it detects aggressive remarks and negative emotional expressions in the text and flags them as abnormal behavior. The input is the transformed text information, and the output is the analysis results and the abnormal behavior flag.

[0735] Step 5:

[0736] If abnormal behavior is detected, the server generates a warning report. This report includes specific details of the detected abnormal behavior and related information. The generated report is sent to an augmented reality display device for visualization. The input is the abnormal behavior flag, and the output is the warning report.

[0737] Step 6:

[0738] The user receives a visually represented warning report through an augmented reality display device. The specific details of abnormal behavior are provided visually, serving as a tool to prompt immediate action. The input is the warning report, and the output is the user's recognition and response.

[0739] 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.

[0740] This invention is a system that detects abnormal behavior and takes appropriate action by analyzing the actions and emotions of individuals in the workplace in real time. In addition to analyzing video and audio data, this system combines an emotion engine to identify the user's emotions, contributing to improved workplace safety and employee well-being.

[0741] Video and audio data analysis

[0742] The server acquires video data from cameras placed in the workplace area and analyzes the behavior using machine learning algorithms. This makes it possible to monitor the behavior of specific individuals in real time and detect signs of abnormal behavior. The server also collects audio data from microphones and converts it into text data. Based on this text data, natural language processing technology is used to identify linguistic anomalies.

[0743] Emotion recognition by an emotion engine

[0744] The server uses an emotion engine to analyze the user's emotional state from video and audio data. The emotion engine detects facial expressions and tone of voice to determine emotions such as joy, anger, and sadness. This enables more accurate detection of abnormal behavior by incorporating emotional data.

[0745] Detection and response to abnormal behavior

[0746] If abnormal behavior or inappropriate emotional states are detected, the server immediately generates an alert report and notifies the relevant administrator or person in charge. This report records details of the behavior and emotions, helping to quickly determine appropriate action. For example, if it is detected that a participant is emotionally agitated and showing discomfort during a meeting, the situation can be reported to the administrator, who can then be instructed to take appropriate action to prevent the problem from escalating.

[0747] Post-meeting evaluation and feedback

[0748] After the meeting ends, the server analyzes data including each participant's behavior and emotions to create individual feedback reports. These reports, delivered to employees via their devices, include trends and areas for behavioral improvement based on their emotional state, promoting self-improvement.

[0749] In this way, by utilizing information about emotional states, the system can simultaneously improve employees' emotional well-being and job performance.

[0750] The following describes the processing flow.

[0751] Step 1:

[0752] The server receives video data in real time from cameras installed in the workplace and sends the data to a sequential processing stream. This prepares the data for real-time analysis.

[0753] Step 2:

[0754] The server applies machine learning models to the received video data to analyze the movements and facial expressions of individual people. It uses face recognition and motion detection algorithms to extract deviations from normal behavioral patterns.

[0755] Step 3:

[0756] The server acquires audio data from microphones installed in the conference room or designated area and stores it as digital audio. The audio processing module then passes this to the speech recognition engine, which converts it into text data.

[0757] Step 4:

[0758] The server analyzes the converted text data using natural language processing algorithms, identifying aggressive or suspicious behavior by analyzing the content and tone of the statements.

[0759] Step 5:

[0760] The server uses an emotion engine to identify emotional states from video and audio data. By analyzing changes in facial features and voice pitch, it classifies the user's emotions into categories such as "joy," "anger," and "sadness."

[0761] Step 6:

[0762] If abnormal behavior or negative emotional states are detected, the server generates an alert report summarizing the situation. This report includes detailed observations and information necessary for action, and is immediately sent to the administrator.

[0763] Step 7:

[0764] The terminal displays suggestions generated during the meeting to participants. These suggestions include specific measures to stimulate discussion based on sentiment data, thereby improving the quality of the meeting.

[0765] Step 8:

[0766] After the meeting, the server analyzes each participant's behavioral and emotional data to perform individual evaluations and generate feedback reports. These reports are distributed to employees via their devices to support their individual growth.

[0767] Step 9:

[0768] Based on the feedback received, users can develop action plans for self-improvement. The feedback serves as a guide for specific behavioral modifications.

[0769] Through these steps, the system comprehensively monitors the work environment, enabling early detection of anomalies and simultaneous improvement of productivity.

[0770] (Example 2)

[0771] 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".

[0772] In the workplace, effectively detecting and promptly responding to abnormal employee behavior and emotional changes is crucial for improving safety and operational efficiency. However, conventional systems are limited to detection based on behavioral and verbal abnormalities, making it difficult to achieve highly accurate detection and immediate response that takes emotional states into account.

[0773] 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.

[0774] In this invention, the server includes means for receiving and analyzing video data to identify a person's movements, means for acquiring audio data and converting it into text data, and means for identifying emotional states from the analyzed data. This enables highly accurate detection of emotional changes and abnormal emotions, and allows for rapid warning reporting and feedback.

[0775] "Video data" refers to visual information acquired from cameras in the workplace environment, and serves as basic data for analyzing people's movements and facial expressions.

[0776] "Means for identifying movements" refers to technologies that analyze video data to identify the movements and behavioral patterns of individual people and detect abnormal movements.

[0777] "Audio data" refers to acoustic signals acquired through a microphone, and is used to analyze conversation content and voice tone.

[0778] "Methods for converting to text data" refers to methods for converting audio data into text data, making the content analyzable.

[0779] A "means for detecting abnormal behavior" refers to a system that identifies unusual behavior or emotional changes based on video and audio data.

[0780] "Means for generating alert reports" refers to a process for documenting details of abnormal behavior or emotions when they are detected and notifying relevant parties.

[0781] "Means for identifying emotional states" refers to technologies that analyze video and audio data to determine a person's emotions (e.g., joy, anger, sadness).

[0782] "Methods for providing feedback" refer to techniques used to show participants areas for improvement and emotional tendencies based on analyzed behavioral and emotional data.

[0783] This invention is a system that analyzes human behavior and emotions in real time and detects anomalies in order to improve safety and employee well-being in the workplace environment. The main components of the invention consist of a camera, microphone, server, and terminal.

[0784] The server processes video data collected by the camera and uses TensorFlow to analyze behavior using machine learning algorithms, including open-source libraries. Through this analysis, the server identifies the behavioral state of specific individuals and detects abnormal behavior. Additionally, audio data acquired by the microphone is converted into text using speech recognition services such as Google Cloud Speech-to-Text, and natural language processing techniques are used to identify linguistic anomalies.

[0785] Furthermore, the server uses an emotion recognition API to analyze emotional states from video and audio. This analysis can detect facial expressions and voice tone, and identify emotions such as joy or anger. The emotion data improves the accuracy of detecting abnormal behavior.

[0786] When an anomaly is detected, an alert report is immediately generated and notified to the appropriate administrator. This report records specific details of the behavior and emotions, facilitating a swift response. For example, if a participant becomes emotionally unstable during a meeting, the system can detect this change and notify the administrator, preventing the problem from escalating.

[0787] Furthermore, after the meeting, the server comprehensively analyzes the participants' behavior and emotions and generates a feedback report. This report includes patterns of emotional changes and points for improving behavior, and is provided to employees via their terminals. This allows users to understand their own emotional tendencies and aim to improve their work performance.

[0788] Examples of prompt messages for this system include the following:

[0789] "Please tell me how to analyze emotional data in the workplace to detect abnormal behavior and suggest improvements."

[0790] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0791] Step 1:

[0792] The server acquires video and audio data in real time from cameras and microphones placed in the workplace. It receives video and audio data streams as input and prepares them as datasets for processing. Specific operations include temporal slicing and buffering of the data.

[0793] Step 2:

[0794] The server uses the acquired video data to analyze human movement by applying machine learning models such as TensorFlow. It uses video frames as input and decodes the action in each frame using a motion recognition algorithm. This process generates action labels as output, enabling the detection of anomalies in the movement of people and objects.

[0795] Step 3:

[0796] The server converts audio data into text data using a speech recognition service such as Google Cloud Speech-to-Text. It takes the audio signal as input, extracts acoustic features, and sequentially converts the audio into text information. The output of this process is conversation data in text format.

[0797] Step 4:

[0798] The server analyzes the converted text data using natural language processing techniques to identify linguistic anomalies. The input is text data, and semantic analysis is performed via a natural language processing engine to detect abnormal language patterns. The output includes alert information for unusual language usage.

[0799] Step 5:

[0800] The server uses an emotion recognition API to perform emotion analysis based on the analyzed video and audio. It takes image features and audio spectra as input and uses a deep learning model to classify emotion labels. The output is generated as a dataset of emotional states, identifying emotions such as joy, anger, and sadness.

[0801] Step 6:

[0802] When abnormal behavior or emotional changes are detected, the server generates an alert report and notifies administrators and other relevant parties who require attention. At this stage, the anomaly detection data is used as input, and a detailed anomaly report is created using a template for structuring the report. The generated output is sent via email or message notification.

[0803] Step 7:

[0804] After the event ends, the server analyzes data for each participant and generates feedback reports. Using behavioral and emotional datasets as input, statistical models and evaluation algorithms are employed to generate reports on emotional tendencies and areas for improvement for each employee. The final output is the report delivered via the terminal.

[0805] This processing flow enables the system to achieve real-time safety in the workplace and improve employee well-being.

[0806] (Application Example 2)

[0807] 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".

[0808] In the workplace, it is crucial to manage worker safety and mental health simultaneously and to detect abnormal behavior and emotional changes early. However, conventional systems lack the means to comprehensively analyze and notify this information in real time, resulting in a challenge in adequately addressing the improvement of workplace safety and worker well-being.

[0809] 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.

[0810] In this invention, the server includes means for receiving and analyzing video data to identify the actions of an individual, means for acquiring audio data and converting it into text data, means for detecting abnormal behavior based on the analyzed video and audio data, means for monitoring the individual's biological information in real time and analyzing their emotional state, and means for sending an alarm to a management device and instructing appropriate action based on the emotional analysis. This enables real-time monitoring of changes in the actions and emotions of workers in the workplace environment, allowing for a rapid response.

[0811] "Video data" refers to digital or analog data representing visual information acquired by a camera or other imaging device.

[0812] "Analysis" is a method of processing collected data to extract specific information or understand its characteristics.

[0813] An "individual" is an object being monitored within a system, usually referring to a human worker, but may also include other living organisms or objects.

[0814] "Action" refers to the physical actions or movements performed by an individual, and usually means bodily movements.

[0815] "Audio data" refers to sound information acquired through devices such as microphones, expressed in digital or analog format.

[0816] "Converting to text data" is the process of analyzing information such as audio and video and converting it into digital format as text.

[0817] "Detection of abnormal behavior" refers to identifying behaviors that deviate from normal behavioral patterns and recognizing the signs of such behavior.

[0818] "Biometric information" refers to data about an individual's physical state and function, including heart rate and respiratory rate.

[0819] "Analysis of emotional state" is a process that aims to evaluate an individual's psychological state and identify specific emotions such as joy or anger.

[0820] "Sending an alarm" is a systemic action to notify relevant parties of information when an anomaly or specific conditions are met.

[0821] "Management device" refers to hardware or software that controls the entire monitoring system and aggregates or distributes necessary information.

[0822] To implement this invention, a system is required in which multiple devices and software components work in cooperation. The server simultaneously acquires video and audio data from cameras and microphones in the workplace environment and analyzes them to identify the actions and emotional states of individuals. Specifically, the video data is processed in real time using an image processing library such as OpenCV, and an emotion recognition engine called EmotionEngine determines emotions from the individual's facial expressions. The audio data is converted into text data using an audio processing library such as librosa.

[0823] If the analysis detects abnormal behavior or emotional abnormalities such as stress, the server sends an alert to the management device, which then notifies the administrator's smartphone or computer. This management device uses a module called AlertManager to generate the alert. Appropriate countermeasures are automatically suggested, enabling a rapid response to the situation.

[0824] As a concrete example, when a new production line starts operation, the system automatically monitors whether employees are experiencing stress due to the change in environment. If an anomaly is detected, managers can use that information to readjust the work environment settings and implement measures to reduce the burden on workers.

[0825] To realize such a system, it is possible to obtain more detailed analysis results by supplying advanced prompt sentences to the generating AI model. For example, a prompt sentence such as, "Please give me ideas for an effective monitoring system to analyze abnormal behavior and emotions in the work environment," could be considered.

[0826] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0827] Step 1:

[0828] The server acquires video data from cameras installed in the workplace environment. It analyzes the received video data using OpenCV to identify individual movements. Specifically, it extracts the coordinates and movement features of individuals and passes this data to subsequent processing steps.

[0829] Step 2:

[0830] The server acquires audio data from microphones in the environment. This audio data is converted into text data using the librosa library. The audio waveform, as input, is analyzed into frequency components, and the content is output in text format through a speech recognition algorithm. This text data is used to detect abnormal behavior.

[0831] Step 3:

[0832] The device uses EmotionEngine to recognize emotional states from analyzed video and audio data. It analyzes facial expressions and tone of voice as input and outputs emotions such as anger and joy. This allows for the identification of signs of abnormal behavior based on emotions.

[0833] Step 4:

[0834] The server detects abnormal behavior based on analyzed behavioral and emotional data. If an abnormality is detected, it sends an alert to the management device using AlertManager. By analyzing the input abnormality information and immediately generating an alert and notifying the administrator's terminal, it encourages a quick response.

[0835] Step 5:

[0836] After receiving an alert as an administrator, the user selects the appropriate course of action. This selection process can be guided by suggestions based on prompts provided by a generative AI model. The user then reviews the output information on the appropriate course of action and applies it in the actual field.

[0837] 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.

[0838] 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.

[0839] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0840] 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.

[0841] 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.

[0842] 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.

[0843] 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.

[0844] 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.

[0845] 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."

[0846] 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.

[0847] 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.

[0848] 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.

[0849] 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.

[0850] 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.

[0851] 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.

[0852] 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.

[0853] 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.

[0854] 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.

[0855] 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.

[0856] 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.

[0857] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0858] The following is further disclosed regarding the embodiments described above.

[0859] (Claim 1)

[0860] A means of receiving video data and analyzing it to identify the movements of a person,

[0861] A means of acquiring audio data and converting it into text data,

[0862] A means for detecting abnormal behavior based on analyzed video and audio data,

[0863] A means for generating and notifying a warning report when abnormal behavior is detected,

[0864] A system that includes this.

[0865] (Claim 2)

[0866] The system according to claim 1, comprising means for monitoring the progress of a meeting and detecting stagnation in the discussion, and for generating and presenting proposals to participants.

[0867] (Claim 3)

[0868] The system according to claim 1, comprising means for evaluating individual job performance and generating feedback reports.

[0869] "Example 1"

[0870] (Claim 1)

[0871] A means for receiving video information, analyzing it, and identifying the action,

[0872] A means for acquiring audio information and converting it into text information,

[0873] A means for detecting abnormal behavior based on analyzed video and audio information,

[0874] A means for generating and notifying a warning report when abnormal behavior is detected,

[0875] A means of monitoring the progress of discussions and making proposals using generative AI,

[0876] A means for evaluating job performance and generating feedback reports,

[0877] A system that includes this.

[0878] (Claim 2)

[0879] The system according to claim 1, which detects a stagnation in the discussion and presents the generated proposals to the participants.

[0880] (Claim 3)

[0881] The system according to claim 1, which records and evaluates individual contributions and generates a detailed improvement report.

[0882] "Application Example 1"

[0883] (Claim 1)

[0884] A means of receiving video information and analyzing it to identify human movements,

[0885] A means for acquiring audio information and converting it into text information,

[0886] A means for detecting abnormal behavior based on analyzed video and audio information,

[0887] A means of generating and notifying a warning report when abnormal behavior is detected,

[0888] A means of visualizing detected abnormal behavior in real time using an augmented reality display device,

[0889] A system that includes this.

[0890] (Claim 2)

[0891] The system according to claim 1, comprising means for monitoring the progress of a meeting and detecting stagnation in the discussion, and for generating and presenting proposals to participants.

[0892] (Claim 3)

[0893] The system according to claim 1, comprising means for evaluating the performance of individual tasks and generating feedback reports.

[0894] "Example 2 of combining an emotion engine"

[0895] (Claim 1)

[0896] A means of receiving video data and analyzing it to identify the movements of a person,

[0897] A means of acquiring audio data and converting it into text data,

[0898] A means for detecting abnormal behavior based on analyzed video and audio data,

[0899] A means for generating and notifying a warning report when abnormal behavior is detected,

[0900] A means of identifying emotional states from analytical data,

[0901] A means of generating reports that prompt immediate action based on emotional changes or abnormal emotions,

[0902] A means of organizing data based on sentiment analysis and providing feedback to participants,

[0903] A system that includes this.

[0904] (Claim 2)

[0905] The system according to claim 1, comprising means for monitoring the progress of a meeting and detecting stagnation in the discussion, and for generating and presenting proposals to participants.

[0906] (Claim 3)

[0907] The system according to claim 1, comprising means for evaluating individual job performance and generating feedback reports.

[0908] "Application example 2 when combining with an emotional engine"

[0909] (Claim 1)

[0910] A means for receiving video data and analyzing it to identify the movements of an individual,

[0911] A means of acquiring audio data and converting it into text data,

[0912] A means for detecting abnormal behavior based on analyzed video and audio data,

[0913] A means for generating and notifying a warning report when abnormal behavior is detected,

[0914] A means of monitoring an individual's biological information in real time and analyzing their emotional state,

[0915] A means of sending an alarm to a management device based on emotion analysis and instructing it to take appropriate action,

[0916] A system that includes this.

[0917] (Claim 2)

[0918] The system according to claim 1, comprising means for monitoring the progress of a meeting and detecting stagnation in the discussion, and for generating and presenting proposals to participants.

[0919] (Claim 3)

[0920] The system according to claim 1, comprising means for evaluating individual work performance and generating feedback reports. [Explanation of symbols]

[0921] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of receiving video data and analyzing it to identify the movements of a person, A means of acquiring audio data and converting it into text data, A means for detecting abnormal behavior based on analyzed video and audio data, A means for generating and notifying a warning report when abnormal behavior is detected, A system that includes this.

2. The system according to claim 1, comprising means for monitoring the progress of a meeting and detecting stagnation in the discussion, and for generating and presenting proposals to participants.

3. The system according to claim 1, comprising means for evaluating the performance of individual jobs and generating feedback reports.

Images