System

Abstract

A system is provided.SOLUTION: A system comprising: means for recording voice information; means for transmitting the recorded voice information to a server; means for transcribing the voice information using a voice-recognition technique in the server; means for analyzing the transcribed information and evaluating logical thinking, eyes, and emotions; means for calculating a matching degree with a company based on the analysis result; and means for notifying users of the result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Conventional interview processes tend to rely on the subjectivity of the interviewer, particularly their experience and skills, resulting in a lack of consistency and objectivity in evaluations. Furthermore, interviews are inefficient because they require a great deal of time and labor costs. Furthermore, there has been no technology available for accurately evaluating candidates using vast amounts of interview data. Therefore, the present invention aims to solve these problems and provide a system that enables efficient and highly accurate interview evaluations. [Means for solving the problem]

[0005] The present invention solves the problems by the following means. First, it provides a means for recording voice data and a means for transmitting the recorded voice data to a server. Next, it provides a means for transcribing the voice data using voice recognition technology on the server, and a means for analyzing the transcribed data and evaluating logical thinking, eye contact, and emotions. Furthermore, it provides a system that includes a means for calculating the degree of match with the company based on the analysis results and a means for notifying the user of the results, thereby improving the consistency and objectivity of interview evaluations.

[0006] In addition, the system includes a means for using an automated conversation agent to analyze voice data, a means for the automated conversation agent to ask questions to applicants and collect voice data, and a means for transcribing and analyzing the exchanges of questions and answers in real time.The system also provides a means for a server to collect and manage a huge amount of interview data, a means for efficiently converting the collected data into a training dataset, and a means for training a generative AI model by improving the network architecture and introducing a new loss function.

[0007] Furthermore, the system provides a means for the server to perform a 360-degree evaluation, calculate the degree of match based on criteria set by the company, and notify company personnel of the analysis results in real time, thereby improving the efficiency and accuracy of the interview process and enabling companies to accurately evaluate the talent they are looking for.

[0008] "Audio data" refers to audio information collected during an interview and stored in the form of a recorded audio file.

[0009] "Server" means a central computer system operating on a network that processes and analyzes audio data and stores and distributes the results.

[0010] A "terminal" is a device used during an interview, and is a device for recording audio data and transmitting it to a server.

[0011] "Transcription" is the process of converting audio data into text using speech recognition technology.

[0012] "Logical thinking" refers to the examinee's logical thinking ability and is an indicator used to evaluate problem-solving ability and logical problem-solving processes.

[0013] "Eye contact" refers to the movement of the eyes during an interview, and is an element that is evaluated by tracking the position and movement of the eyes using video analysis.

[0014] "Emotion" is a factor that evaluates the emotional state of the candidate during the interview, and is determined using emotion analysis technology from voice and facial expressions.

[0015] "Match level" is an indicator that shows how closely the candidate matches the company's required standards based on the analysis results.

[0016] An "automated conversational agent" is an AI chatbot that is used to automatically conduct conversations and collect interview data.

[0017] A "training dataset" is a collection of data used to train a generative AI model, including the collected interview data.

[0018] "Network architecture" refers to the design structure of a model and is a technical framework for improving the performance of generative AI models.

[0019] A "loss function" is a function used to evaluate the error during training of a generative AI model and to improve the model's accuracy.

[0020] "360-degree evaluation" refers to a method of evaluating test takers from multiple angles, and is a process of conducting a comprehensive evaluation based on multiple evaluation criteria.

[0021] "Analysis results" are evaluation information that includes transcription data and the results of logical thinking, perspective, and emotional analysis.

[0022] "Real-time" refers to immediate processing and notification without delay, meaning that evaluation results are instantly delivered to company personnel. [Brief explanation of the drawings]

[0023] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0024] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0025] First, the terms used in the following description will be explained.

[0026] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0027] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0028] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0029] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0030] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0031] [First embodiment]

[0032] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0033] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0034] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0035] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0036] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0037] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0038] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0039] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0040] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0041] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0042] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0043] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0044] The present invention provides a system that aims to improve the efficiency of the interview process and the accuracy of the evaluation. This system automates and optimizes the series of processes including recording, transmitting, transcribing, analyzing, calculating the match degree, and notifying the results of the interview. Specific embodiments and examples are described below.

[0045] System Overview

[0046] The system consists of a "terminal" operated by the "user" (interviewer or applicant) and a "server" that processes and analyzes the voice data. The terminal and server are connected via a network, and various data is sent and received.

[0047] Program processing overview

[0048] 1. Start recording

[0049] When the user (interviewer or applicant) starts the interview, he / she presses the "Start Recording" button on the terminal.

[0050] The device will have its recording function enabled and will record the audio data of the interview.

[0051] 2. Sending audio data

[0052] After the interview is over, the user presses the "End Recording" button on the terminal.

[0053] The device will stop recording and send the saved audio data to the server.

[0054] 3. Transcription

[0055] The server transcribes the received voice data using voice recognition technology.

[0056] The audio data is converted into text data.

[0057] 4. Data Analysis

[0058] The server analyzes the transcribed text data and evaluates the test-taker's logical thinking, gaze, and emotions.

[0059] If necessary, in the case of face-to-face interviews, video data will also be analyzed to analyze eye movements and facial expressions.

[0060] 5. Match Calculation

[0061] The server calculates the degree of match with the company from the analysis results based on the evaluation criteria set by the company.

[0062] The match level is an indicator of how closely the candidate matches the profile of the employee the company is looking for.

[0063] 6. Result notification

[0064] The server transmits the analysis results and the degree of match to the terminal.

[0065] The terminal displays the results on the screen so that the user can check them.

[0066] Specific examples

[0067] Example 1: Conducting an in-person interview

[0068] When the user (interviewer) begins the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop recording. The device sends the recorded data to the server, which transcribes the audio data and performs logical thinking and sentiment analysis. The match degree is calculated based on the analysis results and the results are sent to the device. The user (interviewer) can check the results and download an evaluation report.

[0069] Example 2: Conducting an automated interview

[0070] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes and analyzes the voice in real time. Finally, the analysis results calculate the degree of match based on the required criteria, and the results are sent to the device. The user (applicant) can check the results.

[0071] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0072] The system includes multiple evaluation criteria to fairly evaluate candidates who best fit the skillsets desired by companies. By combining speech recognition technology with generative AI models, it achieves highly accurate transcription and sentiment analysis. Furthermore, the server's distributed computing technology allows for fast and efficient processing of even large data sets, significantly improving the efficiency of the entire interview process.

[0073] Additional information

[0074] In addition, the generative AI model is trained using a vast amount of interview data collected. The training process includes improvements to the network architecture and the introduction of a new loss function to optimize the model's learning speed and accuracy. Furthermore, multifaceted data analysis is incorporated to achieve a 360-degree evaluation of the company's desired talent profile.

[0075] The above is a specific embodiment of the present invention. This system not only significantly improves the efficiency and accuracy of interviews, but also eliminates interviewer bias and achieves fairer evaluations.

[0076] The processing flow will be explained below.

[0077] Step 1:

[0078] The user clicks the Start Interview button. The device activates the recording function and starts recording audio data. The device displays a confirmation message on the screen saying "Recording has started."

[0079] Step 2:

[0080] When the interview is finished, the user clicks the stop recording button. The device stops recording and saves the recorded audio data.

[0081] Step 3:

[0082] The device sends the saved voice data to the server. The server receives the voice data and returns a confirmation message to the device stating, "The voice data has been successfully sent to the server." The device displays this confirmation message on its screen.

[0083] Step 4:

[0084] The server passes the received voice data to the voice recognition engine, which then analyzes and transcribes the voice data. The voice recognition engine analyzes the voice data and converts it into text data.

[0085] Step 5:

[0086] The server passes the transcribed text data to the analysis engine, which then uses the text data to evaluate logical thinking, analyze emotions, and analyze gaze.

[0087] Step 6:

[0088] (For face-to-face interviews)

[0089] The device sends the video data collected during recording to a server, which then passes the video data to an analysis engine for analysis of gaze and facial expressions.

[0090] Step 7:

[0091] The server calculates the degree of match based on the data obtained by the analysis engine and in accordance with the evaluation criteria set by the company.

[0092] Step 8:

[0093] The server sends the match rate and analysis results to the terminal, which displays these results on the screen so that the user can check them.

[0094] Step 9:

[0095] (In the case of automated interviews)

[0096] The user selects "Automatic Interview" in the application. The device sends a request for an automated interview to the server. The server launches an automated conversation agent (AI chatbot) and asks questions to the user.

[0097] Step 10:

[0098] The AI ​​chatbot on the server asks the user predefined questions in sequence and receives answers in voice or text. The server records the exchange of questions and answers and performs real-time speech recognition, transcription, and analysis.

[0099] Step 11:

[0100] The server calculates the analysis results and the degree of match and sends them to the terminal. The user can check the results and download the report if necessary.

[0101] Example 1

[0102] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0103] The current interview process requires a lot of manual work, is inefficient, and the interviewer's subjectivity has a significant impact on the evaluation, making it difficult to achieve fair and accurate evaluation. Furthermore, transcribing and analyzing audio data takes time, making it unsuitable for large-scale data processing. Furthermore, conventional methods have the problem of making it difficult to set evaluation criteria that effectively utilize interview data and calculate the degree of match between candidates.

[0104] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0105] In this invention, the server includes means for starting recording at the start of the interview, means for stopping recording at the end of the interview and transmitting the audio data to the server, means for transcribing the audio data using speech recognition technology on the server, means for analyzing the transcribed data and evaluating logical thinking, line of sight, and emotions, means for calculating the degree of match with the company based on the analysis results, means for notifying the user of the results, and means for displaying the results. This enables the automation and optimization of the interview process, and enables efficient and fair evaluation.

[0106] The "means for starting recording at the start of an interview" is a function for enabling the built-in microphone of the terminal by user operation at the start of an interview and recording audio data.

[0107] "Means for stopping recording at the end of the interview and sending audio data to the server" is a function that stops recording at the end of the interview by user operation and sends the recorded audio data to the server using a secure protocol.

[0108] "Means for transcribing voice data using voice recognition technology on the server" is a function for converting voice data received by the server into text data using voice recognition technology.

[0109] "Means of analyzing transcribed data and evaluating logical thinking, perspective, and emotions" refers to a function that analyzes transcribed text data using a natural language processing model and sentiment analysis API to evaluate the test taker's logical thinking, perspective, and emotions.

[0110] "Means for calculating the degree of match with a company based on the analysis results" is a function for calculating the degree of compatibility between an applicant and a company from the results of data analysis, based on the evaluation criteria set by the company.

[0111] The "means for notifying the user of the results" is a function for sending the calculated analysis results and the degree of match to the user's terminal so that the user can check them.

[0112] The "means for displaying results" is a function for visually displaying the analysis results and the degree of match received on the terminal so that the user can easily check them.

[0113] An "automated conversational agent" is a conversational system such as a chatbot that uses AI, and is a technology for automating voice or text interactions with users.

[0114] "Huge interview data" refers to a collection of large amounts of recorded interview data, and is a variety of data related to interviews, including audio data, text data, video data, and the like.

[0115] A "training dataset" is a collection of data used to train and optimize a generative AI model.

[0116] The present invention provides a system that aims to improve the efficiency of the interview process and the accuracy of the evaluation. This system automates and optimizes the series of processes including recording, transmitting, transcribing, analyzing, calculating the match degree, and notifying the results of the interview. Specific embodiments and examples are described below.

[0117] System Overview

[0118] The system consists of a "terminal" operated by the "user" (interviewer or applicant) and a "server" that processes and analyzes voice data. The terminal and server are connected via a network, and various data is sent and received. The main hardware used includes PCs, tablets, and smartphones. The software used includes a voice recording application, a voice recognition engine, and a natural language processing model.

[0119] Program processing overview

[0120] 1. Start recording

[0121] When the user wants to start the interview, they press the "Start Recording" button on the dedicated application (e.g., "Interview Recorder app") installed on their device.

[0122] The device will activate its built-in microphone and record audio data.

[0123] 2. Sending audio data

[0124] After the interview is over, the user presses the "End Recording" button.

[0125] The device will stop recording and send the saved audio data to the server using a secure protocol (HTTPS).

[0126] 3. Transcription

[0127] The server transcribes the received audio data using the "Google (registered trademark) Cloud Speech-to-Text" API.

[0128] The audio data is converted into text data.

[0129] 4. Data Analysis

[0130] The server analyzes the transcribed text data and evaluates the following:

[0131] An "NLP model" is used to evaluate logical thinking, specifically by analyzing the frequency of keyword occurrences and the logical structure of sentences.

[0132] To evaluate sentiment, we use the Sentiment Analysis API, which calculates positive and negative sentiment indicators from text.

[0133] If necessary, video data is also analyzed to analyze gaze and facial expressions using "OpenCV."

[0134] 5. Match Calculation

[0135] The server calculates the degree of match with the company from the analysis results based on the evaluation criteria set by the company.

[0136] The analysis results are input into a custom recommendation engine to calculate a match score.

[0137] 6. Result notification

[0138] The server transmits the analysis results and the degree of match to the terminal.

[0139] The terminal displays the results on the screen so that the user can check them.

[0140] Specific examples

[0141] Example 1: Conducting an in-person interview

[0142] When the user (interviewer) begins a face-to-face interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop recording. The device sends the recorded data to the server, which transcribes it using "Google Cloud Speech-to-Text." The NLP model then performs text analysis and sentiment analysis, and calculates the degree of match based on criteria set by the company. The analysis results and degree of match are sent to the device, where the user (interviewer) can check the results and download an evaluation report.

[0143] Example 2: Conducting an automated interview

[0144] The user (applicant) activates the automated interview function, and the AI ​​chatbot conducts the interview. The chatbot asks questions one by one, and the user (applicant) responds by voice or text. The server transcribes the responses in real time using Google Cloud Speech-to-Text and analyzes them using an NLP model. Finally, the degree of match is calculated, and the results are sent to the device. The user (applicant) can check the results.

[0145] In this way, by combining voice recognition technology with generative AI models, this system can significantly improve the efficiency and accuracy of interviews. It also enables multifaceted data analysis to select candidates who best fit the skill sets required by companies.

[0146] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0147] Step 1:

[0148] When the user starts the interview, he or she presses the "Start Recording" button on the dedicated application.

[0149] Input: User operation (instruction to start recording)

[0150] The device will activate the built-in microphone and begin recording.

[0151] Output: Recorded audio data

[0152] Step 2:

[0153] The terminal will display the recording status to let the user know that recording is in progress.

[0154] Input: Recording time and status information

[0155] The terminal displays the recording status on the user interface.

[0156] Output: Status display during recording

[0157] Step 3:

[0158] After the interview is over, the user presses the "End Recording" button.

[0159] Input: User operation (instruction to end recording)

[0160] The device will stop recording and temporarily save the audio data.

[0161] Output: Saved audio data

[0162] Step 4:

[0163] The terminal transmits the recorded voice data to the server.

[0164] Input: Saved audio data

[0165] The device sends the audio data to the server using the HTTPS protocol.

[0166] Output: Audio data sent to the server

[0167] Step 5:

[0168] The server receives the audio data and transcribes it using speech recognition technology.

[0169] Input: Received audio data

[0170] The server converts the audio into text data using the Google Cloud Speech-to-Text API.

[0171] Output: Transcribed text data

[0172] Step 6:

[0173] The server analyzes the transcribed text data and evaluates logical thinking, gaze, and emotions.

[0174] Input: Transcribed text data

[0175] The server does the following:

[0176] Evaluating logical thinking using an "NLP model" (analysis of keyword frequency and logical structure of sentences)

[0177] Evaluate sentiment using the Sentiment Analysis API (calculate positive and negative sentiment indicators from text)

[0178] Optionally, use OpenCV to analyze gaze and facial expressions from video data.

[0179] Output: Analysis results

[0180] Step 7:

[0181] The server calculates the degree of match with the company based on the analysis results.

[0182] Input: Analysis results and evaluation criteria set by the company

[0183] The server uses a custom recommendation engine to calculate the match score.

[0184] Output: Match score

[0185] Step 8:

[0186] The server transmits the analysis results and the degree of match to the terminal.

[0187] Input: Analysis results and match scores

[0188] The server sends the analysis results and the degree of match to the terminal.

[0189] Output: Submitted analysis results and match scores

[0190] Step 9:

[0191] The terminal displays the results on the screen so that the user can check them.

[0192] Input: Submitted analysis results and match rate

[0193] The terminal displays the results visually for easy confirmation by the user.

[0194] Output: Displayed analysis results and match scores

[0195] (Application example 1)

[0196] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0197] It is not easy for factory employees to efficiently learn new operations and procedures and improve their proficiency. Conventional methods require managers to manually evaluate each one, which is time-consuming and can lead to inequity. Furthermore, because worker aptitude assessments tend to be subjective, it is difficult to assign appropriate personnel.

[0198] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0199] In this invention, the server includes a means for recording voice data, a means for transmitting the recorded voice data to the server, a means for transcribing the voice data using voice recognition technology in the server, a means for analyzing the transcribed data and evaluating logical thinking, proficiency, and aptitude, a means for calculating the skill level of the worker based on the analysis results, and a means for notifying the user of the results. This automates the aptitude evaluation of workers in the factory, making it possible to improve the efficiency and fairness of the evaluation.

[0200] "Voice data" is digital information that records the voice and voice instructions of the worker.

[0201] "Recording" is the act of saving audio data and making it available for later use.

[0202] A "server" is a computer system that receives, analyzes, and stores voice data via a network.

[0203] "Speech recognition technology" is a technology that converts collected voice data into text.

[0204] "Transcription" is the process of converting audio data into text data.

[0205] "Data analysis" is a method of analyzing transcribed text data according to various criteria.

[0206] "Logical thinking" refers to the ability to think logically and is a skill for solving problems systematically and efficiently.

[0207] "Skill level" is an index that evaluates how skilled a worker is at a particular task or operation.

[0208] "Aptitude" is an indicator of how well a worker is suited to a particular role or task.

[0209] "Skill level" indicates the degree of technical ability and knowledge of a worker.

[0210] "Notification" refers to the act of informing the user of the analysis results and evaluations.

[0211] An "automated conversational agent" is a program that can interact with employees through voice and text.

[0212] A "training dataset" is a collection of data used to train a generative AI model.

[0213] "Network architecture" is the structural design that enables generative AI models to operate efficiently.

[0214] A "loss function" is a function used to minimize error when training a generative AI model.

[0215] MODE FOR CARRYING OUT THE INVENTION

[0216] The present invention provides a system for automating the aptitude evaluation and skill level determination of workers in a factory. Specific embodiments are described below.

[0217] System Overview

[0218] The system consists of a "terminal" operated by the worker and a "server" that processes and analyzes the voice data. The terminal and server are connected via a network, and various data is sent and received.

[0219] Program processing overview

[0220] The system has the following main functions:

[0221] 1. Audio data recording:

[0222] When a worker starts work, he or she presses the "Start Recording" button on the terminal, which then records the work voice and behavioral data.

[0223] 2. Sending audio data:

[0224] When the worker finishes the work and presses the "Stop Recording" button, the terminal stops recording and sends the recorded voice data to the server.

[0225] 3. Transcription:

[0226] The server transcribes the received voice data using voice recognition technology and converts it into text data.

[0227] 4. Data Analysis:

[0228] The server analyzes the converted text data and evaluates the worker's logical thinking, proficiency, and aptitude.

[0229] 5. Skill Level Calculation:

[0230] The server calculates the worker's skill level based on the evaluation results.

[0231] 6. Result notification:

[0232] The server transmits the evaluation result and skill level to the terminal, which displays them.

[0233] Hardware and software used

[0234] Hardware:

[0235] Terminal: Equipped with a microphone for recording audio data and buttons for starting and stopping recording.

[0236] Server: Equipped with a high-performance processor and large memory capacity, it can quickly analyze voice data.

[0237] software:

[0238] Speech recognition technology: Transcribing voice data using Google's speech recognition API.

[0239] Data Analysis: Use the Transformers library to analyze the sentiment, logical thinking, proficiency, and aptitude of the transcribed data.

[0240] Specific examples

[0241] For example, a worker in charge of explaining how to operate a new machine presses the start recording button on the terminal to begin the explanation. When the work is finished, the worker presses the stop recording button to stop recording the audio data, and the terminal sends the data to the server. The server transcribes the audio data and analyzes the text data to evaluate the worker's proficiency and aptitude, and sends the results to the terminal. This allows managers to check the evaluation results of the workers and provide appropriate guidance and placement.

[0242] Prompt Sentence Examples

[0243] Below are some example prompts that can be used as input to generative AI models:

[0244] "Transcribe the audio data, analyze the employee's logical thinking and proficiency, and calculate their aptitude."

[0245] This will allow for quick and efficient evaluation of workers in your factory.

[0246] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0247] Step 1:

[0248] Processing content: The user presses the "Start recording" button on the device.

[0249] Specific operation: Before starting work in the factory, the user presses the "Start Recording" button to enable the terminal's recording function.

[0250] Input and Output: The input is the user pressing a button, and the output is the device starting recording.

[0251] Step 2:

[0252] Processing details: The device records the work audio.

[0253] Specific operation: The terminal starts recording the worker's voice data using the microphone. The voice data being recorded is saved in a buffer in real time.

[0254] Input and output: The input is the worker's voice, and the voice data is stored in the terminal's buffer memory.

[0255] Step 3:

[0256] Processing content: The user presses the "Stop Recording" button on the device.

[0257] Specific operation: When the work is finished, the user presses the "End Recording" button on the device to stop recording.

[0258] Input and Output: Input is the user pressing a button, and output is the device stopping recording and saving the audio data to a file.

[0259] Step 4:

[0260] Processing content: The device sends voice data to the server.

[0261] Specific operation: The device sends the saved voice data to the server via the network. The data transfer uses the HTTP protocol.

[0262] Input and Output: The input is the audio data stored on the device, and the output is the audio file sent to the server.

[0263] Step 5:

[0264] Processing: The server transcribes the audio data.

[0265] Specific operation: The server converts the received voice data into text using speech recognition technology. It uses Google's speech recognition API.

[0266] Input and Output: The input is the audio data sent to the server, and the output is the transcribed text data.

[0267] Step 6:

[0268] Processing content: The server analyzes the text data.

[0269] Specific operation: The server uses the Transformers library to analyze the transcribed text data for sentiment, logical thinking, proficiency, and aptitude.

[0270] Input and output: The input is transcribed text data, and the output is the analytical results (emotions, logical thinking, proficiency, aptitude).

[0271] Step 7:

[0272] Processing content: The server calculates the skill level.

[0273] Specific operation: Based on the analysis results, the server calculates the worker's skill level. Calculating the skill level includes the process of comparing the analysis results with the evaluation criteria.

[0274] Inputs and Outputs: The inputs are the parameters of the analysis results, and the output is the calculated skill level.

[0275] Step 8:

[0276] Processing content: The server sends the evaluation results to the terminal.

[0277] Specific operation: The server sends the calculated skill level and evaluation results to the terminal. Data is transferred using the HTTP protocol.

[0278] Input and Output: The input is the skill level and the evaluation result, and the output is the evaluation result data sent to the terminal.

[0279] Step 9:

[0280] Processing content: The terminal displays the evaluation results to the user.

[0281] Specific operation: The terminal displays the received evaluation results on the user interface, and the user can view the evaluation results on the terminal screen.

[0282] Input and output: The input is the evaluation result data sent from the server, and the output is the evaluation result displayed on the terminal screen.

[0283] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0284] The present invention is a system for improving the efficiency and fairness of the interview process, and in particular incorporates an "emotion engine" that evaluates the user's emotions by comprehensively analyzing audio data and video data. Specific embodiments and their operation are described below.

[0285] System configuration

[0286] The system consists of a "terminal" that records and transmits voice data, a "server" that analyzes voice and video data, and an "emotion engine" that recognizes the user's emotions. These components work together while sending and receiving data via a network.

[0287] Program processing overview

[0288] 1. Start recording

[0289] The user clicks the Start Interview button. The device activates its audio and video recording functions and records audio and video data.

[0290] The device will display a confirmation message saying "Recording has started."

[0291] 2. Transmission of audio and video data

[0292] When the interview is finished, the user clicks the end button, and the device stops recording and saves the audio and video data.

[0293] The device sends the stored audio and video data to the server, which returns a confirmation message to the device, which displays it.

[0294] 3. Transcription and data analysis

[0295] The server passes the received voice data to a voice recognition engine to perform transcription.

[0296] The server passes the transcribed text data to an analysis engine, which evaluates logical thinking, analyzes emotions, and analyzes gaze.

[0297] 4. Operation of the Emotion Engine

[0298] The emotion engine of the server analyzes the user's emotion from the audio and video data.

[0299] Emotional data is analyzed in real time and incorporated into logical thinking and other assessments.

[0300] The server calculates the degree of match with the company, including the results of sentiment analysis.

[0301] 5. Notification of Results

[0302] The server sends the analysis results and the degree of match to the terminal, which then displays the results on the screen.

[0303] Users can review the results and download the report if desired.

[0304] Specific examples

[0305] Example 1: Conducting an in-person interview

[0306] When the user (interviewer or applicant) starts the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop the audio and video recording. The device sends the audio and video data to the server, which transcribes the audio data and analyzes emotions using an emotion engine. The evaluation results and match degree are calculated and sent to the device. The user (interviewer or applicant) can check the results and download the evaluation report.

[0307] Example 2: Conducting an automated interview

[0308] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes the voice in real time, analyzes emotions using an emotion engine, and analyzes logical thinking and eye contact. Finally, the analysis results are calculated along with the degree of match and sent to the device. The user (applicant) can check the results.

[0309] Emotion Engine Details

[0310] The emotion engine uses advanced machine learning algorithms to analyze a user's emotional state from audio and video data. Audio analysis uses features such as tone, speed, and pitch of the voice. Video analysis, on the other hand, evaluates facial expressions, eye contact, and other physical movements. The emotion engine integrates these data in real time to accurately recognize the user's emotional state.

[0311] Additional information

[0312] This system integrates voice recognition and emotion analysis technologies to improve the objectivity and fairness of interviews. Furthermore, it utilizes distributed computing technology on the server to efficiently process large amounts of interview data. This significantly improves the efficiency and accuracy of interviews, enabling companies to select more suitable candidates.

[0313] The above is a specific embodiment of the present invention. This system comprehensively analyzes audio data and video data to achieve highly accurate emotion evaluation, thereby contributing to the improvement of the interview process.

[0314] The processing flow will be explained below.

[0315] Step 1:

[0316] The user clicks the Start Interview button. The device activates the audio and video recording functions and begins recording audio and video data during the interview. The device displays a confirmation message on the screen saying "Recording has started."

[0317] Step 2:

[0318] When the interview is finished, the user clicks the end button, and the device stops recording audio and video, saving the recorded audio and video data.

[0319] Step 3:

[0320] The device sends the stored audio and video data to the server. The server receives the audio and video data and returns a confirmation message to the device stating, "Audio and video data has been successfully sent to the server." The device displays this confirmation message on its screen.

[0321] Step 4:

[0322] The server passes the received voice data to the voice recognition engine, which begins analyzing and transcribing the voice data. The voice recognition engine analyzes the voice data and converts it into text data. After conversion is complete, the server receives the text data.

[0323] Step 5:

[0324] The server then passes the video data to an analysis engine along with the transcribed text data, which then analyzes the user's gaze and facial expressions. The video analysis engine then analyzes the video data and extracts information about the user's gaze and facial expressions.

[0325] Step 6:

[0326] The server passes the text data, eye movement data, and facial expression data obtained from the analysis engine to the emotion engine, which analyzes the user's emotions using the tone, pitch, and speed of the voice and the facial expression information extracted from the video data.

[0327] Step 7:

[0328] The server's emotion engine synthesizes the analyzed emotion data and generates a rich emotion profile, which can be used as an indicator of the user's emotional state.

[0329] Step 8:

[0330] The server calculates the degree of match based on all analysis results, including the emotional profile, according to the evaluation criteria set by the company, which includes a multifaceted evaluation of logical thinking, eye contact, emotions, and voice content.

[0331] Step 9:

[0332] The server sends the analysis results and the degree of match to the device, which displays the results on the screen for the user to review. The user can then review the results and download a report if necessary.

[0333] Step 10:

[0334] (In the case of automated interviews)

[0335] The user selects "Automatic Interview" in the application. The device sends a request for an automated interview to the server. The server launches an automated conversation agent (AI chatbot) and asks the user the defined questions.

[0336] Step 11:

[0337] The AI ​​chatbot on the server asks the user questions one after another and receives answers via voice or text. The server records the exchange of questions and answers, performs real-time speech recognition and transcription, and also analyzes the user's emotions using an emotion engine.

[0338] Step 12:

[0339] The server evaluates all data obtained in real time and calculates the degree of match. Finally, the analysis results are sent to the device along with the degree of match. The user can check the results and download the report if necessary.

[0340] Example 2

[0341] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0342] The current interview process relies on a lot of manual work and subjective judgment, resulting in a lack of efficiency and fairness. In particular, because it relies on the interviewer's subjectivity, evaluation criteria are vague and emotions and logical thinking may not be properly evaluated. In addition, managing and analyzing large amounts of interview data is time-consuming, placing a heavy burden on companies. This often makes it difficult to select the right candidates.

[0343] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0344] In this invention, the server includes means for recording audio data and video data, means for transmitting the recorded audio data and video data to the server, means for transcribing the audio data using speech recognition technology in the server, means for analyzing the transcribed data and video data and evaluating logical thinking, line of sight, and emotions, means for calculating the degree of match with the company based on the analysis results, and means for notifying the user of the degree of match with the company and the analysis results, thereby making it possible to improve the efficiency and fairness of the interview process.

[0345] "Audio data" refers to voice information recorded during an interview or conversation that has been stored in digital form.

[0346] "Video data" refers to image and video information recorded during an interview or conversation that has been saved in digital format.

[0347] A "server" is a remote computer system used for data storage, analysis, and management.

[0348] "Voice recognition technology" is a technology for converting voice input into text data.

[0349] "Transcription" is the process of converting audio data into text data.

[0350] "Logical thinking" is an evaluation criterion that refers to the ability to think logically and reason.

[0351] "Gaze" refers to the direction and movement of the gaze, and refers to gaze behavior during an interview, etc.

[0352] "Emotions" refers to the psychological state judged from facial expressions, tone of voice, etc. during the interview.

[0353] "Evaluation" is the process of quantifying or grading the user's abilities and emotions based on the analyzed data.

[0354] "Match level" is an indicator of the compatibility between the interviewee and the company.

[0355] "User" refers to an interviewer or applicant who uses the system.

[0356] "Notification" refers to the act of transmitting information by the server to inform the user of the analysis results and evaluations.

[0357] The present invention provides a system for comprehensively analyzing audio data and video data to improve the efficiency and fairness of the interview process. Specific embodiments of this system will be described below.

[0358] System configuration

[0359] The system consists of three main components:

[0360] 1. "Devices" that record audio and video data

[0361] 2. The "server" that analyzes the received data and calculates the results

[0362] 3. "Emotion Engine" that analyzes emotional states

[0363] These components operate in cooperation with each other via a network.

[0364] Hardware and Software Use

[0365] Terminal

[0366] When the user starts the interview, the device will record audio and video data using the camera and microphone. The device also has audio and video recording functions to save the audio and video data. When the recording is finished, the data is sent to the server.

[0367] server

[0368] When the server receives the voice data, it transcribes it using the Google Cloud Speech-to-Text API. The resulting text data is then analyzed using IBM Watson® Natural Language Understanding. The analysis includes logical thinking, eye gaze analysis, and sentiment analysis. Finally, the server calculates the match rate with the company based on the analysis results and sends the results to the device.

[0369] Emotion Engine

[0370] The server's emotion engine uses the Emotion API to analyze the user's emotional state from audio and video data. Audio analysis uses features such as tone, speed, and pitch of the voice. Video analysis evaluates facial expressions, eye movements, and other body movements.

[0371] Specific examples

[0372] Example 1: Conducting an in-person interview

[0373] When the user (interviewer or applicant) starts the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop the audio and video recording. The device sends the audio and video data to the server, which transcribes the audio data and analyzes emotions using an emotion engine. The evaluation results and match degree are calculated and sent to the device. The user (interviewer or applicant) can check the results and download the evaluation report.

[0374] Example 2: Conducting an automated interview

[0375] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes the voice in real time, analyzes emotions using an emotion engine, and analyzes logical thinking and eye contact. Finally, the analysis results are calculated along with the degree of match and sent to the device. The user (applicant) can check the results.

[0376] Prompt Sentence Examples

[0377] "Tell me about a system that can streamline and improve fairness in the interview process for applicants. How can you analyze emotions using audio and video data? Can you also give me some specific use cases?"

[0378] The above is an embodiment of the present invention. By using this system, it is possible to significantly improve the efficiency and fairness of interviews.

[0379] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0380] Step 1:

[0381] The user clicks the start interview button. The device responds by enabling the camera and microphone and starting to record audio and video data. The input here is the user's click action, and the output is the audio and video data that has started to be recorded. Specifically, pressing the record button initializes the device's camera and microphone.

[0382] Step 2:

[0383] The device records the audio and video data of the interview in real time. The recorded data is temporarily stored in the device's local storage. The input at this time is the user's voice and movements, and the output is the audio and video data stored in the local storage. Specifically, audio data is obtained from the microphone input, and video data is obtained from the camera input.

[0384] Step 3:

[0385] When the interview is over, the user clicks the end button. The device stops recording and saves the recorded data completely. The input at this time is the click of the end button, and the output is the final audio and video data file. Specifically, the recording is stopped and the data is saved.

[0386] Step 4:

[0387] The device sends the stored audio and video data to the server. The input here is the audio and video data stored in the device's local storage, and the output is the data sent to the server. Specifically, the data is uploaded to the server via the network using the HTTP or HTTPS protocol.

[0388] Step 5:

[0389] The server checks the received data and passes the audio data to the Google Cloud Speech-to-Text API for transcription. The input is the audio data received by the server, and the output is the transcribed text data. Specifically, the API is called to generate text data from the audio data.

[0390] Step 6:

[0391] The server passes the transcribed text data to IBM Watson Natural Language Understanding to evaluate logical thinking. The input here is the transcribed text data, and the output is the evaluation results. Specifically, the text data is analyzed and logical thinking ability is quantified.

[0392] Step 7:

[0393] At the same time, the server passes the video data to the Emotion API to analyze the emotional state. The input is the video data, and the output is the user's emotional state data. Specifically, the system analyzes facial expressions and gaze from the video data to recognize emotions.

[0394] Step 8:

[0395] The server integrates the logical thinking evaluation results and emotional state data to calculate the match degree with the company. The inputs are the evaluation results and emotional state data, and the output is a match degree score. Specifically, the server calculates the match degree by weighting each evaluation data.

[0396] Step 9:

[0397] The server sends the analysis results and match scores to the terminal. The input at this time is the analysis results and match scores at the server, and the output is notification data to the terminal. Specifically, the analysis results and evaluation scores are transferred to the terminal via the network.

[0398] Step 10:

[0399] The terminal displays the received analysis results and match scores on the screen. Finally, the user can check the results and download the evaluation report if necessary. The input in this case is the notification data from the server, and the output is the displayed analysis results and a downloadable evaluation report. Specifically, the user interface is updated and the results are displayed.

[0400] (Application example 2)

[0401] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0402] Conventional interview systems analyze only voice data, making it difficult to fully grasp the interviewer's emotions and intentions. Furthermore, there are insufficient means to improve the fairness and efficiency of interviews. It is also difficult to grasp customer emotions and provide appropriate service when serving customers in brick-and-mortar stores. Conventional technologies cannot analyze customer emotions in real time and suggest optimal customer service methods based on that analysis, making it difficult to improve customer satisfaction.

[0403] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0404] In this invention, the server includes a means for recording voice data, a means for transmitting the recorded voice data to the server, a means for transcribing the voice data using voice recognition technology on the server, a means for analyzing the transcribed data and evaluating logical thinking, gaze, and emotions, a means for calculating the degree of match with the company based on the analysis results, and a means for notifying the user of the results. This enables the interviewer's emotions to be recognized with high accuracy from voice and video data during the interview, enabling fair and efficient evaluation. Furthermore, in customer service at physical stores, customer satisfaction can be improved by analyzing customer emotions in real time using smart devices equipped with cameras and microphones and proposing optimal customer service methods based on the analysis results.

[0405] "Audio data" is a recording of sound collected through an audio input device such as a microphone.

[0406] A "server" is a computer system that transmits, receives, and processes data over a network.

[0407] "Voice recognition technology" is a technology that analyzes voice data and converts it into text format.

[0408] "Transcription" is the process of recognizing audio data and converting it into corresponding text data.

[0409] "Logical thinking" is the process of evaluating logical thinking.

[0410] "Gaze" is the process of assessing visual direction and focus.

[0411] "Emotion" is the process of assessing a user's mental state from audio and video data.

[0412] "Company match" is an indicator that evaluates the compatibility between the interviewer and the company.

[0413] "Notification" is the process of informing users of the analysis results.

[0414] A "camera" is a device that captures video data.

[0415] A "microphone" is a device that collects sound.

[0416] A "display device" is a device that visually presents analysis results and proposals.

[0417] The "Emotion Engine" is a system that analyzes emotions from audio and video data.

[0418] "Feedback" is the process of returning analysis results and suggestions to users.

[0419] An "automated conversational agent" is a system that uses artificial intelligence to automatically conduct dialogue.

[0420] A "smart device" is an electronic device with advanced features that can connect to the Internet.

[0421] A "generative AI model" is an artificial intelligence model that is trained using large amounts of data.

[0422] A "prompt" is a textual instruction that is input to a generative AI model.

[0423] The present invention provides a system for evaluating emotions by analyzing audio and video data during interviews and customer service, thereby realizing efficient and fair dialogue. Specific embodiments of the system are described below.

[0424] System configuration

[0425] The system consists of the following main components:

[0426] 1. Device: This includes smart devices equipped with cameras and microphones (e.g., smart glasses, smartphones, and head-mounted displays). These devices capture audio and video data in real time and have communication capabilities.

[0427] 2. Server: A cloud server for analyzing data, equipped with speech recognition technology, an emotion engine, and a data analysis engine. For example, Amazon Web Services (AWS (registered trademark)) or Google Cloud Platform (GCP) can be used.

[0428] 3. Emotion engine: Software for analyzing emotions from audio and video data. Microsoft® Azure® Emotion API and Affectiva SDK are used.

[0429] 4. Generative AI model: An artificial intelligence model trained using large amounts of training data, which then generates appropriate feedback and suggestions.

[0430] Program processing

[0431] 1. Data capture and transmission:

[0432] A user (e.g., a store clerk) wears a terminal such as smart glasses and captures audio and video data.

[0433] The captured data is sent to a server in real time.

[0434] 2. Data Analysis:

[0435] The server transcribes the received voice data using a speech recognition engine, such as Google Cloud Speech-to-Text or IBM Watson Speech to Text.

[0436] Once transcribed, the text data is passed to an emotion engine that analyzes the emotions of the customer or interviewer. Voice analysis uses features such as tone, speed, and pitch, while video analysis evaluates facial expressions, eye movements, and other physical movements.

[0437] 3. Real-time feedback:

[0438] Based on the analysis results of the emotion engine, the server proposes optimal actions to the user (e.g., a store clerk) in real time, enabling the user to communicate efficiently with customers and interviewers.

[0439] The suggestions are fed back to the terminal's display device.

[0440] Specific examples

[0441] For example, imagine a salesperson wearing smart glasses explaining a luxury product to a customer. In this case, the glasses will analyze the customer's facial expressions and tone of voice in real time to gauge their level of interest or anxiety, and based on that, the glasses will suggest specific actions such as providing a more detailed explanation or offering a topic to help them relax.

[0442] Specific examples of prompts are as follows:

[0443] "Read the following sentence in a fully emotive tone: 'This watch is a classic design from the 1950s and is worn by many celebrities.'"

[0444] Using this prompt, the generative AI model can generate dialogue in specific scenarios and create data to improve the accuracy of its sentiment analysis.

[0445] The above is a specific embodiment of the present invention. This system realizes highly accurate emotion evaluation through integrated analysis of audio and video data, enabling efficient and fair operation in interviews and brick-and-mortar stores.

[0446] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0447] Step 1:

[0448] A user (e.g., a store clerk) puts on smart glasses and begins serving customers or conducting an interview. The smart glasses have a built-in camera and microphone to capture audio and video data. Pressing the device's activation button starts audio and video recording.

[0449] Input: User's audio and video

[0450] Output: Captured audio and video data

[0451] Step 2:

[0452] The device transmits the captured audio and video data to the server in real time, and during the transmission, the device temporarily stores the data to ensure network stability.

[0453] Input: Captured audio and video data

[0454] Output: Sending audio and video data to the server

[0455] Step 3:

[0456] The server then passes the received voice data to a speech recognition engine, such as Google Cloud Speech-to-Text or IBM Watson Speech to Text, which analyzes the voice data and generates corresponding text data.

[0457] Input: Audio data

[0458] Output: Transcribed text data

[0459] Step 4:

[0460] The server passes the transcribed text data and video data to the emotion engine for emotion analysis. The emotion engine extracts features from the audio and video data and analyzes emotions using a machine learning model.

[0461] Input: Transcribed text and video data

[0462] Output: Emotion analysis results

[0463] Step 5:

[0464] Based on the results of the sentiment analysis, the server generates the optimal action using a generative AI model, which combines past data with prompts to create specific suggestions for the user.

[0465] Input: Sentiment analysis results

[0466] Output: Recommendations for optimal action

[0467] Example prompt sentence:

[0468] "Read the following sentence in a fully emotive tone: 'This watch is a classic design from the 1950s and is worn by many celebrities.'"

[0469] Step 6:

[0470] The server sends the generated suggestions to the display device of the terminal and provides feedback to the user, who then sees the suggestions in real time on the display of the smart glasses.

[0471] Input: Generated proposals

[0472] Output: Feedback to a display device

[0473] Step 7:

[0474] The user takes appropriate action based on the displayed suggestions, which are then captured again by the camera and microphone and reflected in the next cycle.

[0475] Input: Shown suggestions

[0476] Output: User actions

[0477] The above are the specific steps of the processing of the program of the system that realizes the application example.

[0478] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0479] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0480] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0481] [Second embodiment]

[0482] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0483] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0484] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0485] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0486] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0487] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0488] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0489] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0490] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0491] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0492] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0493] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0494] The present invention provides a system that aims to improve the efficiency of the interview process and the accuracy of the evaluation. This system automates and optimizes the series of processes including recording, transmitting, transcribing, analyzing, calculating the match degree, and notifying the results of the interview. Specific embodiments and examples are described below.

[0495] System Overview

[0496] The system consists of a "terminal" operated by the "user" (interviewer or applicant) and a "server" that processes and analyzes the voice data. The terminal and server are connected via a network, and various data is sent and received.

[0497] Program processing overview

[0498] 1. Start recording

[0499] When the user (interviewer or applicant) starts the interview, he / she presses the "Start Recording" button on the terminal.

[0500] The device will have its recording function enabled and will record the audio data of the interview.

[0501] 2. Sending audio data

[0502] After the interview is over, the user presses the "End Recording" button on the terminal.

[0503] The device will stop recording and send the saved audio data to the server.

[0504] 3. Transcription

[0505] The server transcribes the received voice data using voice recognition technology.

[0506] The audio data is converted into text data.

[0507] 4. Data Analysis

[0508] The server analyzes the transcribed text data and evaluates the test-taker's logical thinking, gaze, and emotions.

[0509] If necessary, in the case of face-to-face interviews, video data will also be analyzed to analyze eye movements and facial expressions.

[0510] 5. Match Calculation

[0511] The server calculates the degree of match with the company from the analysis results based on the evaluation criteria set by the company.

[0512] The match level is an indicator of how closely the candidate matches the profile of the employee the company is looking for.

[0513] 6. Result notification

[0514] The server transmits the analysis results and the degree of match to the terminal.

[0515] The terminal displays the results on the screen so that the user can check them.

[0516] Specific examples

[0517] Example 1: Conducting an in-person interview

[0518] When the user (interviewer) begins the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop recording. The device sends the recorded data to the server, which transcribes the audio data and performs logical thinking and sentiment analysis. The match degree is calculated based on the analysis results and the results are sent to the device. The user (interviewer) can check the results and download an evaluation report.

[0519] Example 2: Conducting an automated interview

[0520] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes and analyzes the voice in real time. Finally, the analysis results calculate the degree of match based on the required criteria, and the results are sent to the device. The user (applicant) can check the results.

[0521] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0522] The system includes multiple evaluation criteria to fairly evaluate candidates who best fit the skillsets desired by companies. By combining speech recognition technology with generative AI models, it achieves highly accurate transcription and sentiment analysis. Furthermore, the server's distributed computing technology allows for fast and efficient processing of even large data sets, significantly improving the efficiency of the entire interview process.

[0523] Additional information

[0524] In addition, the generative AI model is trained using a vast amount of interview data collected. The training process includes improvements to the network architecture and the introduction of a new loss function to optimize the model's learning speed and accuracy. Furthermore, multifaceted data analysis is incorporated to achieve a 360-degree evaluation of the company's desired talent profile.

[0525] The above is a specific embodiment of the present invention. This system not only significantly improves the efficiency and accuracy of interviews, but also eliminates interviewer bias and achieves fairer evaluations.

[0526] The processing flow will be explained below.

[0527] Step 1:

[0528] The user clicks the Start Interview button. The device activates the recording function and starts recording audio data. The device displays a confirmation message on the screen saying "Recording has started."

[0529] Step 2:

[0530] When the interview is finished, the user clicks the stop recording button. The device stops recording and saves the recorded audio data.

[0531] Step 3:

[0532] The device sends the saved voice data to the server. The server receives the voice data and returns a confirmation message to the device stating, "The voice data has been successfully sent to the server." The device displays this confirmation message on its screen.

[0533] Step 4:

[0534] The server passes the received voice data to the voice recognition engine, which then analyzes and transcribes the voice data. The voice recognition engine analyzes the voice data and converts it into text data.

[0535] Step 5:

[0536] The server passes the transcribed text data to the analysis engine, which then uses the text data to evaluate logical thinking, analyze emotions, and analyze gaze.

[0537] Step 6:

[0538] (For face-to-face interviews)

[0539] The device sends the video data collected during recording to a server, which then passes the video data to an analysis engine for analysis of gaze and facial expressions.

[0540] Step 7:

[0541] The server calculates the degree of match based on the data obtained by the analysis engine and in accordance with the evaluation criteria set by the company.

[0542] Step 8:

[0543] The server sends the match rate and analysis results to the terminal, which displays these results on the screen so that the user can check them.

[0544] Step 9:

[0545] (In the case of automated interviews)

[0546] The user selects "Automatic Interview" in the application. The device sends a request for an automated interview to the server. The server launches an automated conversation agent (AI chatbot) and asks questions to the user.

[0547] Step 10:

[0548] The AI ​​chatbot on the server asks the user predefined questions in sequence and receives answers in voice or text. The server records the exchange of questions and answers and performs real-time speech recognition, transcription, and analysis.

[0549] Step 11:

[0550] The server calculates the analysis results and the degree of match and sends them to the terminal. The user can check the results and download the report if necessary.

[0551] Example 1

[0552] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0553] The current interview process requires a lot of manual work, is inefficient, and the interviewer's subjectivity has a significant impact on the evaluation, making it difficult to achieve fair and accurate evaluation. Furthermore, transcribing and analyzing audio data takes time, making it unsuitable for large-scale data processing. Furthermore, conventional methods have the problem of making it difficult to set evaluation criteria that effectively utilize interview data and calculate the degree of match between candidates.

[0554] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0555] In this invention, the server includes means for starting recording at the start of the interview, means for stopping recording at the end of the interview and transmitting the audio data to the server, means for transcribing the audio data using speech recognition technology on the server, means for analyzing the transcribed data and evaluating logical thinking, line of sight, and emotions, means for calculating the degree of match with the company based on the analysis results, means for notifying the user of the results, and means for displaying the results. This enables the automation and optimization of the interview process, and enables efficient and fair evaluation.

[0556] The "means for starting recording at the start of an interview" is a function for enabling the built-in microphone of the terminal by user operation at the start of an interview and recording audio data.

[0557] "Means for stopping recording at the end of the interview and sending audio data to the server" is a function that stops recording at the end of the interview by user operation and sends the recorded audio data to the server using a secure protocol.

[0558] "Means for transcribing voice data using voice recognition technology on the server" is a function for converting voice data received by the server into text data using voice recognition technology.

[0559] "Means of analyzing transcribed data and evaluating logical thinking, perspective, and emotions" refers to a function that analyzes transcribed text data using a natural language processing model and sentiment analysis API to evaluate the test taker's logical thinking, perspective, and emotions.

[0560] "Means for calculating the degree of match with a company based on the analysis results" is a function for calculating the degree of compatibility between an applicant and a company from the results of data analysis, based on the evaluation criteria set by the company.

[0561] The "means for notifying the user of the results" is a function for sending the calculated analysis results and the degree of match to the user's terminal so that the user can check them.

[0562] The "means for displaying results" is a function for visually displaying the analysis results and the degree of match received on the terminal so that the user can easily check them.

[0563] An "automated conversational agent" is a conversational system such as a chatbot that uses AI, and is a technology for automating voice or text interactions with users.

[0564] "Huge interview data" refers to a collection of large amounts of recorded interview data, and is a variety of data related to interviews, including audio data, text data, video data, and the like.

[0565] A "training dataset" is a collection of data used to train and optimize a generative AI model.

[0566] The present invention provides a system that aims to improve the efficiency of the interview process and the accuracy of the evaluation. This system automates and optimizes the series of processes including recording, transmitting, transcribing, analyzing, calculating the match degree, and notifying the results of the interview. Specific embodiments and examples are described below.

[0567] System Overview

[0568] The system consists of a "terminal" operated by the "user" (interviewer or applicant) and a "server" that processes and analyzes voice data. The terminal and server are connected via a network, and various data is sent and received. The main hardware used includes PCs, tablets, and smartphones. The software used includes a voice recording application, a voice recognition engine, and a natural language processing model.

[0569] Program processing overview

[0570] 1. Start recording

[0571] When the user wants to start the interview, they press the "Start Recording" button on the dedicated application (e.g., "Interview Recorder app") installed on their device.

[0572] The device will activate its built-in microphone and record audio data.

[0573] 2. Sending audio data

[0574] After the interview is over, the user presses the "End Recording" button.

[0575] The device will stop recording and send the saved audio data to the server using a secure protocol (HTTPS).

[0576] 3. Transcription

[0577] The server transcribes the received audio data using the Google Cloud Speech-to-Text API.

[0578] The audio data is converted into text data.

[0579] 4. Data Analysis

[0580] The server analyzes the transcribed text data and evaluates the following:

[0581] An "NLP model" is used to evaluate logical thinking, specifically by analyzing the frequency of keyword occurrences and the logical structure of sentences.

[0582] To evaluate sentiment, we use the Sentiment Analysis API, which calculates positive and negative sentiment indicators from text.

[0583] If necessary, video data is also analyzed to analyze gaze and facial expressions using "OpenCV."

[0584] 5. Match Calculation

[0585] The server calculates the degree of match with the company from the analysis results based on the evaluation criteria set by the company.

[0586] The analysis results are input into a custom recommendation engine to calculate a match score.

[0587] 6. Result notification

[0588] The server transmits the analysis results and the degree of match to the terminal.

[0589] The terminal displays the results on the screen so that the user can check them.

[0590] Specific examples

[0591] Example 1: Conducting an in-person interview

[0592] When the user (interviewer) begins a face-to-face interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop recording. The device sends the recorded data to the server, which transcribes it using "Google Cloud Speech-to-Text." The NLP model then performs text analysis and sentiment analysis, and calculates the degree of match based on criteria set by the company. The analysis results and degree of match are sent to the device, where the user (interviewer) can check the results and download an evaluation report.

[0593] Example 2: Conducting an automated interview

[0594] The user (applicant) activates the automated interview function, and the AI ​​chatbot conducts the interview. The chatbot asks questions one by one, and the user (applicant) responds by voice or text. The server transcribes the responses in real time using Google Cloud Speech-to-Text and analyzes them using an NLP model. Finally, the degree of match is calculated, and the results are sent to the device. The user (applicant) can check the results.

[0595] In this way, by combining voice recognition technology with generative AI models, this system can significantly improve the efficiency and accuracy of interviews. It also enables multifaceted data analysis to select candidates who best fit the skill sets required by companies.

[0596] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0597] Step 1:

[0598] When the user starts the interview, he or she presses the "Start Recording" button on the dedicated application.

[0599] Input: User operation (instruction to start recording)

[0600] The device will activate the built-in microphone and begin recording.

[0601] Output: Recorded audio data

[0602] Step 2:

[0603] The terminal will display the recording status to let the user know that recording is in progress.

[0604] Input: Recording time and status information

[0605] The terminal displays the recording status on the user interface.

[0606] Output: Status display during recording

[0607] Step 3:

[0608] After the interview is over, the user presses the "End Recording" button.

[0609] Input: User operation (instruction to end recording)

[0610] The device will stop recording and temporarily save the audio data.

[0611] Output: Saved audio data

[0612] Step 4:

[0613] The terminal transmits the recorded voice data to the server.

[0614] Input: Saved audio data

[0615] The device sends the audio data to the server using the HTTPS protocol.

[0616] Output: Audio data sent to the server

[0617] Step 5:

[0618] The server receives the audio data and transcribes it using speech recognition technology.

[0619] Input: Received audio data

[0620] The server converts the audio into text data using the Google Cloud Speech-to-Text API.

[0621] Output: Transcribed text data

[0622] Step 6:

[0623] The server analyzes the transcribed text data and evaluates logical thinking, gaze, and emotions.

[0624] Input: Transcribed text data

[0625] The server does the following:

[0626] Evaluating logical thinking using an "NLP model" (analysis of keyword frequency and logical structure of sentences)

[0627] Evaluate sentiment using the Sentiment Analysis API (calculate positive and negative sentiment indicators from text)

[0628] Optionally, use OpenCV to analyze gaze and facial expressions from video data.

[0629] Output: Analysis results

[0630] Step 7:

[0631] The server calculates the degree of match with the company based on the analysis results.

[0632] Input: Analysis results and evaluation criteria set by the company

[0633] The server uses a custom recommendation engine to calculate the match score.

[0634] Output: Match score

[0635] Step 8:

[0636] The server transmits the analysis results and the degree of match to the terminal.

[0637] Input: Analysis results and match scores

[0638] The server sends the analysis results and the degree of match to the terminal.

[0639] Output: Submitted analysis results and match scores

[0640] Step 9:

[0641] The terminal displays the results on the screen so that the user can check them.

[0642] Input: Submitted analysis results and match rate

[0643] The terminal displays the results visually for easy confirmation by the user.

[0644] Output: Displayed analysis results and match scores

[0645] (Application example 1)

[0646] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0647] It is not easy for factory employees to efficiently learn new operations and procedures and improve their proficiency. Conventional methods require managers to manually evaluate each one, which is time-consuming and can lead to inequity. Furthermore, because worker aptitude assessments tend to be subjective, it is difficult to assign appropriate personnel.

[0648] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0649] In this invention, the server includes a means for recording voice data, a means for transmitting the recorded voice data to the server, a means for transcribing the voice data using voice recognition technology in the server, a means for analyzing the transcribed data and evaluating logical thinking, proficiency, and aptitude, a means for calculating the skill level of the worker based on the analysis results, and a means for notifying the user of the results. This automates the aptitude evaluation of workers in the factory, making it possible to improve the efficiency and fairness of the evaluation.

[0650] "Voice data" is digital information that records the voice and voice instructions of the worker.

[0651] "Recording" is the act of saving audio data and making it available for later use.

[0652] A "server" is a computer system that receives, analyzes, and stores voice data via a network.

[0653] "Speech recognition technology" is a technology that converts collected voice data into text.

[0654] "Transcription" is the process of converting audio data into text data.

[0655] "Data analysis" is a method of analyzing transcribed text data according to various criteria.

[0656] "Logical thinking" refers to the ability to think logically and is a skill for solving problems systematically and efficiently.

[0657] "Skill level" is an index that evaluates how skilled a worker is at a particular task or operation.

[0658] "Aptitude" is an indicator of how well a worker is suited to a particular role or task.

[0659] "Skill level" indicates the degree of technical ability and knowledge of a worker.

[0660] "Notification" refers to the act of informing the user of the analysis results and evaluations.

[0661] An "automated conversational agent" is a program that can interact with employees through voice and text.

[0662] A "training dataset" is a collection of data used to train a generative AI model.

[0663] "Network architecture" is the structural design that enables generative AI models to operate efficiently.

[0664] A "loss function" is a function used to minimize error when training a generative AI model.

[0665] MODE FOR CARRYING OUT THE INVENTION

[0666] The present invention provides a system for automating the aptitude evaluation and skill level determination of workers in a factory. Specific embodiments are described below.

[0667] System Overview

[0668] The system consists of a "terminal" operated by the worker and a "server" that processes and analyzes the voice data. The terminal and server are connected via a network, and various data is sent and received.

[0669] Program processing overview

[0670] The system has the following main functions:

[0671] 1. Audio data recording:

[0672] When a worker starts work, he or she presses the "Start Recording" button on the terminal, which then records the work voice and behavioral data.

[0673] 2. Sending audio data:

[0674] When the worker finishes the work and presses the "Stop Recording" button, the terminal stops recording and sends the recorded voice data to the server.

[0675] 3. Transcription:

[0676] The server transcribes the received voice data using voice recognition technology and converts it into text data.

[0677] 4. Data Analysis:

[0678] The server analyzes the converted text data and evaluates the worker's logical thinking, proficiency, and aptitude.

[0679] 5. Skill Level Calculation:

[0680] The server calculates the worker's skill level based on the evaluation results.

[0681] 6. Result notification:

[0682] The server transmits the evaluation result and skill level to the terminal, which displays them.

[0683] Hardware and software used

[0684] Hardware:

[0685] Terminal: Equipped with a microphone for recording audio data and buttons for starting and stopping recording.

[0686] Server: Equipped with a high-performance processor and large memory capacity, it can quickly analyze voice data.

[0687] software:

[0688] Speech recognition technology: Transcribing voice data using Google's speech recognition API.

[0689] Data Analysis: Use the Transformers library to analyze the sentiment, logical thinking, proficiency, and aptitude of the transcribed data.

[0690] Specific examples

[0691] For example, a worker in charge of explaining how to operate a new machine presses the start recording button on the terminal to begin the explanation. When the work is finished, the worker presses the stop recording button to stop recording the audio data, and the terminal sends the data to the server. The server transcribes the audio data and analyzes the text data to evaluate the worker's proficiency and aptitude, and sends the results to the terminal. This allows managers to check the evaluation results of the workers and provide appropriate guidance and placement.

[0692] Prompt Sentence Examples

[0693] Below are some example prompts that can be used as input to generative AI models:

[0694] "Transcribe the audio data, analyze the employee's logical thinking and proficiency, and calculate their aptitude."

[0695] This will allow for quick and efficient evaluation of workers in your factory.

[0696] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0697] Step 1:

[0698] Processing content: The user presses the "Start recording" button on the device.

[0699] Specific operation: Before starting work in the factory, the user presses the "Start Recording" button to enable the terminal's recording function.

[0700] Input and Output: The input is the user pressing a button, and the output is the device starting recording.

[0701] Step 2:

[0702] Processing details: The device records the work audio.

[0703] Specific operation: The terminal starts recording the worker's voice data using the microphone. The voice data being recorded is saved in a buffer in real time.

[0704] Input and output: The input is the worker's voice, and the voice data is stored in the terminal's buffer memory.

[0705] Step 3:

[0706] Processing content: The user presses the "Stop Recording" button on the device.

[0707] Specific operation: When the work is finished, the user presses the "End Recording" button on the device to stop recording.

[0708] Input and Output: Input is the user pressing a button, and output is the device stopping recording and saving the audio data to a file.

[0709] Step 4:

[0710] Processing content: The device sends voice data to the server.

[0711] Specific operation: The device sends the saved voice data to the server via the network. The data transfer uses the HTTP protocol.

[0712] Input and Output: The input is the audio data stored on the device, and the output is the audio file sent to the server.

[0713] Step 5:

[0714] Processing: The server transcribes the audio data.

[0715] Specific operation: The server converts the received voice data into text using speech recognition technology. It uses Google's speech recognition API.

[0716] Input and Output: The input is the audio data sent to the server, and the output is the transcribed text data.

[0717] Step 6:

[0718] Processing content: The server analyzes the text data.

[0719] Specific operation: The server uses the Transformers library to analyze the transcribed text data for sentiment, logical thinking, proficiency, and aptitude.

[0720] Input and output: The input is transcribed text data, and the output is the analytical results (emotions, logical thinking, proficiency, aptitude).

[0721] Step 7:

[0722] Processing content: The server calculates the skill level.

[0723] Specific operation: Based on the analysis results, the server calculates the worker's skill level. Calculating the skill level includes the process of comparing the analysis results with the evaluation criteria.

[0724] Inputs and Outputs: The inputs are the parameters of the analysis results, and the output is the calculated skill level.

[0725] Step 8:

[0726] Processing content: The server sends the evaluation results to the terminal.

[0727] Specific operation: The server sends the calculated skill level and evaluation results to the terminal. Data is transferred using the HTTP protocol.

[0728] Input and Output: The input is the skill level and the evaluation result, and the output is the evaluation result data sent to the terminal.

[0729] Step 9:

[0730] Processing content: The terminal displays the evaluation results to the user.

[0731] Specific operation: The terminal displays the received evaluation results on the user interface, and the user can view the evaluation results on the terminal screen.

[0732] Input and output: The input is the evaluation result data sent from the server, and the output is the evaluation result displayed on the terminal screen.

[0733] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0734] The present invention is a system for improving the efficiency and fairness of the interview process, and in particular incorporates an "emotion engine" that evaluates the user's emotions by comprehensively analyzing audio data and video data. Specific embodiments and their operation are described below.

[0735] System configuration

[0736] The system consists of a "terminal" that records and transmits voice data, a "server" that analyzes voice and video data, and an "emotion engine" that recognizes the user's emotions. These components work together while sending and receiving data via a network.

[0737] Program processing overview

[0738] 1. Start recording

[0739] The user clicks the Start Interview button. The device activates its audio and video recording functions and records audio and video data.

[0740] The device will display a confirmation message saying "Recording has started."

[0741] 2. Transmission of audio and video data

[0742] When the interview is finished, the user clicks the end button, and the device stops recording and saves the audio and video data.

[0743] The device sends the stored audio and video data to the server, which returns a confirmation message to the device, which displays it.

[0744] 3. Transcription and data analysis

[0745] The server passes the received voice data to a voice recognition engine to perform transcription.

[0746] The server passes the transcribed text data to an analysis engine, which evaluates logical thinking, analyzes emotions, and analyzes gaze.

[0747] 4. Operation of the Emotion Engine

[0748] The emotion engine of the server analyzes the user's emotion from the audio and video data.

[0749] Emotional data is analyzed in real time and incorporated into logical thinking and other assessments.

[0750] The server calculates the degree of match with the company, including the results of sentiment analysis.

[0751] 5. Notification of Results

[0752] The server sends the analysis results and the degree of match to the terminal, which then displays the results on the screen.

[0753] Users can review the results and download the report if desired.

[0754] Specific examples

[0755] Example 1: Conducting an in-person interview

[0756] When the user (interviewer or applicant) starts the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop the audio and video recording. The device sends the audio and video data to the server, which transcribes the audio data and analyzes emotions using an emotion engine. The evaluation results and match degree are calculated and sent to the device. The user (interviewer or applicant) can check the results and download the evaluation report.

[0757] Example 2: Conducting an automated interview

[0758] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes the voice in real time, analyzes emotions using an emotion engine, and analyzes logical thinking and eye contact. Finally, the analysis results are calculated along with the degree of match and sent to the device. The user (applicant) can check the results.

[0759] Emotion Engine Details

[0760] The emotion engine uses advanced machine learning algorithms to analyze a user's emotional state from audio and video data. Audio analysis uses features such as tone, speed, and pitch of the voice. Video analysis, on the other hand, evaluates facial expressions, eye contact, and other physical movements. The emotion engine integrates these data in real time to accurately recognize the user's emotional state.

[0761] Additional information

[0762] This system integrates voice recognition and emotion analysis technologies to improve the objectivity and fairness of interviews. Furthermore, it utilizes distributed computing technology on the server to efficiently process large amounts of interview data. This significantly improves the efficiency and accuracy of interviews, enabling companies to select more suitable candidates.

[0763] The above is a specific embodiment of the present invention. This system comprehensively analyzes audio data and video data to achieve highly accurate emotion evaluation, thereby contributing to the improvement of the interview process.

[0764] The processing flow will be explained below.

[0765] Step 1:

[0766] The user clicks the Start Interview button. The device activates the audio and video recording functions and begins recording audio and video data during the interview. The device displays a confirmation message on the screen saying "Recording has started."

[0767] Step 2:

[0768] When the interview is finished, the user clicks the end button, and the device stops recording audio and video, saving the recorded audio and video data.

[0769] Step 3:

[0770] The device sends the stored audio and video data to the server. The server receives the audio and video data and returns a confirmation message to the device stating, "Audio and video data has been successfully sent to the server." The device displays this confirmation message on its screen.

[0771] Step 4:

[0772] The server passes the received voice data to the voice recognition engine, which begins analyzing and transcribing the voice data. The voice recognition engine analyzes the voice data and converts it into text data. After conversion is complete, the server receives the text data.

[0773] Step 5:

[0774] The server then passes the video data to an analysis engine along with the transcribed text data, which then analyzes the user's gaze and facial expressions. The video analysis engine then analyzes the video data and extracts information about the user's gaze and facial expressions.

[0775] Step 6:

[0776] The server passes the text data, eye movement data, and facial expression data obtained from the analysis engine to the emotion engine, which analyzes the user's emotions using the tone, pitch, and speed of the voice and the facial expression information extracted from the video data.

[0777] Step 7:

[0778] The server's emotion engine synthesizes the analyzed emotion data and generates a rich emotion profile, which can be used as an indicator of the user's emotional state.

[0779] Step 8:

[0780] The server calculates the degree of match based on all analysis results, including the emotional profile, according to the evaluation criteria set by the company, which includes a multifaceted evaluation of logical thinking, eye contact, emotions, and voice content.

[0781] Step 9:

[0782] The server sends the analysis results and the degree of match to the device, which displays the results on the screen for the user to review. The user can then review the results and download a report if necessary.

[0783] Step 10:

[0784] (In the case of automated interviews)

[0785] The user selects "Automatic Interview" in the application. The device sends a request for an automated interview to the server. The server launches an automated conversation agent (AI chatbot) and asks the user the defined questions.

[0786] Step 11:

[0787] The AI ​​chatbot on the server asks the user questions one after another and receives answers via voice or text. The server records the exchange of questions and answers, performs real-time speech recognition and transcription, and also analyzes the user's emotions using an emotion engine.

[0788] Step 12:

[0789] The server evaluates all data obtained in real time and calculates the degree of match. Finally, the analysis results are sent to the device along with the degree of match. The user can check the results and download the report if necessary.

[0790] Example 2

[0791] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0792] The current interview process relies on a lot of manual work and subjective judgment, resulting in a lack of efficiency and fairness. In particular, because it relies on the interviewer's subjectivity, evaluation criteria are vague and emotions and logical thinking may not be properly evaluated. In addition, managing and analyzing large amounts of interview data is time-consuming, placing a heavy burden on companies. This often makes it difficult to select the right candidates.

[0793] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0794] In this invention, the server includes means for recording audio data and video data, means for transmitting the recorded audio data and video data to the server, means for transcribing the audio data using speech recognition technology in the server, means for analyzing the transcribed data and video data and evaluating logical thinking, line of sight, and emotions, means for calculating the degree of match with the company based on the analysis results, and means for notifying the user of the degree of match with the company and the analysis results, thereby making it possible to improve the efficiency and fairness of the interview process.

[0795] "Audio data" refers to voice information recorded during an interview or conversation that has been stored in digital form.

[0796] "Video data" refers to image and video information recorded during an interview or conversation that has been saved in digital format.

[0797] A "server" is a remote computer system used for data storage, analysis, and management.

[0798] "Voice recognition technology" is a technology for converting voice input into text data.

[0799] "Transcription" is the process of converting audio data into text data.

[0800] "Logical thinking" is an evaluation criterion that refers to the ability to think logically and reason.

[0801] "Gaze" refers to the direction and movement of the gaze, and refers to gaze behavior during an interview, etc.

[0802] "Emotions" refers to the psychological state judged from facial expressions, tone of voice, etc. during the interview.

[0803] "Evaluation" is the process of quantifying or grading the user's abilities and emotions based on the analyzed data.

[0804] "Match level" is an indicator of the compatibility between the interviewee and the company.

[0805] "User" refers to an interviewer or applicant who uses the system.

[0806] "Notification" refers to the act of transmitting information by the server to inform the user of the analysis results and evaluations.

[0807] The present invention provides a system for comprehensively analyzing audio data and video data to improve the efficiency and fairness of the interview process. Specific embodiments of this system will be described below.

[0808] System configuration

[0809] The system consists of three main components:

[0810] 1. "Devices" that record audio and video data

[0811] 2. The "server" that analyzes the received data and calculates the results

[0812] 3. "Emotion Engine" that analyzes emotional states

[0813] These components operate in cooperation with each other via a network.

[0814] Hardware and Software Use

[0815] Terminal

[0816] When the user starts the interview, the device will record audio and video data using the camera and microphone. The device also has audio and video recording functions to save the audio and video data. When the recording is finished, the data is sent to the server.

[0817] server

[0818] When the server receives the voice data, it transcribes it using the Google Cloud Speech-to-Text API. The resulting text data is then analyzed using IBM Watson Natural Language Understanding. The analysis includes logical thinking, eye gaze analysis, and sentiment analysis. Finally, the server calculates the match rate with the company based on the analysis results and sends the results to the device.

[0819] Emotion Engine

[0820] The server's emotion engine uses the Emotion API to analyze the user's emotional state from audio and video data. Audio analysis uses features such as tone, speed, and pitch of the voice. Video analysis evaluates facial expressions, eye movements, and other body movements.

[0821] Specific examples

[0822] Example 1: Conducting an in-person interview

[0823] When the user (interviewer or applicant) starts the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop the audio and video recording. The device sends the audio and video data to the server, which transcribes the audio data and analyzes emotions using an emotion engine. The evaluation results and match degree are calculated and sent to the device. The user (interviewer or applicant) can check the results and download the evaluation report.

[0824] Example 2: Conducting an automated interview

[0825] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes the voice in real time, analyzes emotions using an emotion engine, and analyzes logical thinking and eye contact. Finally, the analysis results are calculated along with the degree of match and sent to the device. The user (applicant) can check the results.

[0826] Prompt Sentence Examples

[0827] "Tell me about a system that can streamline and improve fairness in the interview process for applicants. How can you analyze emotions using audio and video data? Can you also give me some specific use cases?"

[0828] The above is an embodiment of the present invention. By using this system, it is possible to significantly improve the efficiency and fairness of interviews.

[0829] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0830] Step 1:

[0831] The user clicks the start interview button. The device responds by enabling the camera and microphone and starting to record audio and video data. The input here is the user's click action, and the output is the audio and video data that has started to be recorded. Specifically, pressing the record button initializes the device's camera and microphone.

[0832] Step 2:

[0833] The device records the audio and video data of the interview in real time. The recorded data is temporarily stored in the device's local storage. The input at this time is the user's voice and movements, and the output is the audio and video data stored in the local storage. Specifically, audio data is obtained from the microphone input, and video data is obtained from the camera input.

[0834] Step 3:

[0835] When the interview is over, the user clicks the end button. The device stops recording and saves the recorded data completely. The input at this time is the click of the end button, and the output is the final audio and video data file. Specifically, the recording is stopped and the data is saved.

[0836] Step 4:

[0837] The device sends the stored audio and video data to the server. The input here is the audio and video data stored in the device's local storage, and the output is the data sent to the server. Specifically, the data is uploaded to the server via the network using the HTTP or HTTPS protocol.

[0838] Step 5:

[0839] The server checks the received data and passes the audio data to the Google Cloud Speech-to-Text API for transcription. The input is the audio data received by the server, and the output is the transcribed text data. Specifically, the API is called to generate text data from the audio data.

[0840] Step 6:

[0841] The server passes the transcribed text data to IBM Watson Natural Language Understanding to evaluate logical thinking. The input here is the transcribed text data, and the output is the evaluation results. Specifically, the text data is analyzed and logical thinking ability is quantified.

[0842] Step 7:

[0843] At the same time, the server passes the video data to the Emotion API to analyze the emotional state. The input is the video data, and the output is the user's emotional state data. Specifically, the system analyzes facial expressions and gaze from the video data to recognize emotions.

[0844] Step 8:

[0845] The server integrates the logical thinking evaluation results and emotional state data to calculate the match degree with the company. The inputs are the evaluation results and emotional state data, and the output is a match degree score. Specifically, the server calculates the match degree by weighting each evaluation data.

[0846] Step 9:

[0847] The server sends the analysis results and match scores to the terminal. The input at this time is the analysis results and match scores at the server, and the output is notification data to the terminal. Specifically, the analysis results and evaluation scores are transferred to the terminal via the network.

[0848] Step 10:

[0849] The terminal displays the received analysis results and match scores on the screen. Finally, the user can check the results and download the evaluation report if necessary. The input in this case is the notification data from the server, and the output is the displayed analysis results and a downloadable evaluation report. Specifically, the user interface is updated and the results are displayed.

[0850] (Application example 2)

[0851] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0852] Conventional interview systems analyze only voice data, making it difficult to fully grasp the interviewer's emotions and intentions. Furthermore, there are insufficient means to improve the fairness and efficiency of interviews. It is also difficult to grasp customer emotions and provide appropriate service when serving customers in brick-and-mortar stores. Conventional technologies cannot analyze customer emotions in real time and suggest optimal customer service methods based on that analysis, making it difficult to improve customer satisfaction.

[0853] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0854] In this invention, the server includes a means for recording voice data, a means for transmitting the recorded voice data to the server, a means for transcribing the voice data using voice recognition technology on the server, a means for analyzing the transcribed data and evaluating logical thinking, gaze, and emotions, a means for calculating the degree of match with the company based on the analysis results, and a means for notifying the user of the results. This enables the interviewer's emotions to be recognized with high accuracy from voice and video data during the interview, enabling fair and efficient evaluation. Furthermore, in customer service at physical stores, customer satisfaction can be improved by analyzing customer emotions in real time using smart devices equipped with cameras and microphones and proposing optimal customer service methods based on the analysis results.

[0855] "Audio data" is a recording of sound collected through an audio input device such as a microphone.

[0856] A "server" is a computer system that transmits, receives, and processes data over a network.

[0857] "Voice recognition technology" is a technology that analyzes voice data and converts it into text format.

[0858] "Transcription" is the process of recognizing audio data and converting it into corresponding text data.

[0859] "Logical thinking" is the process of evaluating logical thinking.

[0860] "Gaze" is the process of assessing visual direction and focus.

[0861] "Emotion" is the process of assessing a user's mental state from audio and video data.

[0862] "Company match" is an indicator that evaluates the compatibility between the interviewer and the company.

[0863] "Notification" is the process of informing users of the analysis results.

[0864] A "camera" is a device that captures video data.

[0865] A "microphone" is a device that collects sound.

[0866] A "display device" is a device that visually presents analysis results and proposals.

[0867] The "Emotion Engine" is a system that analyzes emotions from audio and video data.

[0868] "Feedback" is the process of returning analysis results and suggestions to users.

[0869] An "automated conversational agent" is a system that uses artificial intelligence to automatically conduct dialogue.

[0870] A "smart device" is an electronic device with advanced features that can connect to the Internet.

[0871] A "generative AI model" is an artificial intelligence model that is trained using large amounts of data.

[0872] A "prompt" is a textual instruction that is input to a generative AI model.

[0873] The present invention provides a system for evaluating emotions by analyzing audio and video data during interviews and customer service, thereby realizing efficient and fair dialogue. Specific embodiments of the system are described below.

[0874] System configuration

[0875] The system consists of the following main components:

[0876] 1. Device: This includes smart devices equipped with cameras and microphones (e.g., smart glasses, smartphones, and head-mounted displays). These devices capture audio and video data in real time and have communication capabilities.

[0877] 2. Server: A cloud server for analyzing data, equipped with speech recognition technology, an emotion engine, and a data analysis engine. For example, Amazon Web Services (AWS) or Google Cloud Platform (GCP) are used.

[0878] 3. Emotion Engine: Software for analyzing emotions from audio and video data. Microsoft Azure Emotion API and Affectiva SDK are used.

[0879] 4. Generative AI model: An artificial intelligence model trained using large amounts of training data, which then generates appropriate feedback and suggestions.

[0880] Program processing

[0881] 1. Data capture and transmission:

[0882] A user (e.g., a store clerk) wears a terminal such as smart glasses and captures audio and video data.

[0883] The captured data is sent to a server in real time.

[0884] 2. Data Analysis:

[0885] The server transcribes the received voice data using a speech recognition engine, such as Google Cloud Speech-to-Text or IBM Watson Speech to Text.

[0886] Once transcribed, the text data is passed to an emotion engine that analyzes the emotions of the customer or interviewer. Voice analysis uses features such as tone, speed, and pitch, while video analysis evaluates facial expressions, eye movements, and other physical movements.

[0887] 3. Real-time feedback:

[0888] Based on the analysis results of the emotion engine, the server proposes optimal actions to the user (e.g., a store clerk) in real time, enabling the user to communicate efficiently with customers and interviewers.

[0889] The suggestions are fed back to the terminal's display device.

[0890] Specific examples

[0891] For example, imagine a salesperson wearing smart glasses explaining a luxury product to a customer. In this case, the glasses will analyze the customer's facial expressions and tone of voice in real time to gauge their level of interest or anxiety, and based on that, the glasses will suggest specific actions such as providing a more detailed explanation or offering a topic to help them relax.

[0892] Specific examples of prompts are as follows:

[0893] "Read the following sentence in a fully emotive tone: 'This watch is a classic design from the 1950s and is worn by many celebrities.'"

[0894] Using this prompt, the generative AI model can generate dialogue in specific scenarios and create data to improve the accuracy of its sentiment analysis.

[0895] The above is a specific embodiment of the present invention. This system realizes highly accurate emotion evaluation through integrated analysis of audio and video data, enabling efficient and fair operation in interviews and brick-and-mortar stores.

[0896] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0897] Step 1:

[0898] A user (e.g., a store clerk) puts on smart glasses and begins serving customers or conducting an interview. The smart glasses have a built-in camera and microphone to capture audio and video data. Pressing the device's activation button starts audio and video recording.

[0899] Input: User's audio and video

[0900] Output: Captured audio and video data

[0901] Step 2:

[0902] The device transmits the captured audio and video data to the server in real time, and during the transmission, the device temporarily stores the data to ensure network stability.

[0903] Input: Captured audio and video data

[0904] Output: Sending audio and video data to the server

[0905] Step 3:

[0906] The server then passes the received voice data to a speech recognition engine, such as Google Cloud Speech-to-Text or IBM Watson Speech to Text, which analyzes the voice data and generates corresponding text data.

[0907] Input: Audio data

[0908] Output: Transcribed text data

[0909] Step 4:

[0910] The server passes the transcribed text data and video data to the emotion engine for emotion analysis. The emotion engine extracts features from the audio and video data and analyzes emotions using a machine learning model.

[0911] Input: Transcribed text and video data

[0912] Output: Emotion analysis results

[0913] Step 5:

[0914] Based on the results of the sentiment analysis, the server generates the optimal action using a generative AI model, which combines past data with prompts to create specific suggestions for the user.

[0915] Input: Sentiment analysis results

[0916] Output: Recommendations for optimal action

[0917] Example prompt sentence:

[0918] "Read the following sentence in a fully emotive tone: 'This watch is a classic design from the 1950s and is worn by many celebrities.'"

[0919] Step 6:

[0920] The server sends the generated suggestions to the display device of the terminal and provides feedback to the user, who then sees the suggestions in real time on the display of the smart glasses.

[0921] Input: Generated proposals

[0922] Output: Feedback to a display device

[0923] Step 7:

[0924] The user takes appropriate action based on the displayed suggestions, which are then captured again by the camera and microphone and reflected in the next cycle.

[0925] Input: Shown suggestions

[0926] Output: User actions

[0927] The above are the specific steps of the processing of the program of the system that realizes the application example.

[0928] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0929] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0930] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0931] [Third embodiment]

[0932] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0933] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0934] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0935] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0936] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0937] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0938] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0939] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0940] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0941] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0942] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0943] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0944] The present invention provides a system that aims to improve the efficiency of the interview process and the accuracy of the evaluation. This system automates and optimizes the series of processes including recording, transmitting, transcribing, analyzing, calculating the match degree, and notifying the results of the interview. Specific embodiments and examples are described below.

[0945] System Overview

[0946] The system consists of a "terminal" operated by the "user" (interviewer or applicant) and a "server" that processes and analyzes the voice data. The terminal and server are connected via a network, and various data is sent and received.

[0947] Program processing overview

[0948] 1. Start recording

[0949] When the user (interviewer or applicant) starts the interview, he / she presses the "Start Recording" button on the terminal.

[0950] The device will have its recording function enabled and will record the audio data of the interview.

[0951] 2. Sending audio data

[0952] After the interview is over, the user presses the "End Recording" button on the terminal.

[0953] The device will stop recording and send the saved audio data to the server.

[0954] 3. Transcription

[0955] The server transcribes the received voice data using voice recognition technology.

[0956] The audio data is converted into text data.

[0957] 4. Data Analysis

[0958] The server analyzes the transcribed text data and evaluates the test-taker's logical thinking, gaze, and emotions.

[0959] If necessary, in the case of face-to-face interviews, video data will also be analyzed to analyze eye movements and facial expressions.

[0960] 5. Match Calculation

[0961] The server calculates the degree of match with the company from the analysis results based on the evaluation criteria set by the company.

[0962] The match level is an indicator of how closely the candidate matches the profile of the employee the company is looking for.

[0963] 6. Result notification

[0964] The server transmits the analysis results and the degree of match to the terminal.

[0965] The terminal displays the results on the screen so that the user can check them.

[0966] Specific examples

[0967] Example 1: Conducting an in-person interview

[0968] When the user (interviewer) begins the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop recording. The device sends the recorded data to the server, which transcribes the audio data and performs logical thinking and sentiment analysis. The match degree is calculated based on the analysis results and the results are sent to the device. The user (interviewer) can check the results and download an evaluation report.

[0969] Example 2: Conducting an automated interview

[0970] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes and analyzes the voice in real time. Finally, the analysis results calculate the degree of match based on the required criteria, and the results are sent to the device. The user (applicant) can check the results.

[0971] DETAILED DESCRIPTION OF THE EMBODIMENTS

[0972] The system includes multiple evaluation criteria to fairly evaluate candidates who best fit the skillsets desired by companies. By combining speech recognition technology with generative AI models, it achieves highly accurate transcription and sentiment analysis. Furthermore, the server's distributed computing technology allows for fast and efficient processing of even large data sets, significantly improving the efficiency of the entire interview process.

[0973] Additional information

[0974] In addition, the generative AI model is trained using a vast amount of interview data collected. The training process includes improvements to the network architecture and the introduction of a new loss function to optimize the model's learning speed and accuracy. Furthermore, multifaceted data analysis is incorporated to achieve a 360-degree evaluation of the company's desired talent profile.

[0975] The above is a specific embodiment of the present invention. This system not only significantly improves the efficiency and accuracy of interviews, but also eliminates interviewer bias and achieves fairer evaluations.

[0976] The processing flow will be explained below.

[0977] Step 1:

[0978] The user clicks the Start Interview button. The device activates the recording function and starts recording audio data. The device displays a confirmation message on the screen saying "Recording has started."

[0979] Step 2:

[0980] When the interview is finished, the user clicks the stop recording button. The device stops recording and saves the recorded audio data.

[0981] Step 3:

[0982] The device sends the saved voice data to the server. The server receives the voice data and returns a confirmation message to the device stating, "The voice data has been successfully sent to the server." The device displays this confirmation message on its screen.

[0983] Step 4:

[0984] The server passes the received voice data to the voice recognition engine, which then analyzes and transcribes the voice data. The voice recognition engine analyzes the voice data and converts it into text data.

[0985] Step 5:

[0986] The server passes the transcribed text data to the analysis engine, which then uses the text data to evaluate logical thinking, analyze emotions, and analyze gaze.

[0987] Step 6:

[0988] (For face-to-face interviews)

[0989] The device sends the video data collected during recording to a server, which then passes the video data to an analysis engine for analysis of gaze and facial expressions.

[0990] Step 7:

[0991] The server calculates the degree of match based on the data obtained by the analysis engine and in accordance with the evaluation criteria set by the company.

[0992] Step 8:

[0993] The server sends the match rate and analysis results to the terminal, which displays these results on the screen so that the user can check them.

[0994] Step 9:

[0995] (In the case of automated interviews)

[0996] The user selects "Automatic Interview" in the application. The device sends a request for an automated interview to the server. The server launches an automated conversation agent (AI chatbot) and asks questions to the user.

[0997] Step 10:

[0998] The AI ​​chatbot on the server asks the user predefined questions in sequence and receives answers in voice or text. The server records the exchange of questions and answers and performs real-time speech recognition, transcription, and analysis.

[0999] Step 11:

[1000] The server calculates the analysis results and the degree of match and sends them to the terminal. The user can check the results and download the report if necessary.

[1001] Example 1

[1002] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1003] The current interview process requires a lot of manual work, is inefficient, and the interviewer's subjectivity has a significant impact on the evaluation, making it difficult to achieve fair and accurate evaluation. Furthermore, transcribing and analyzing audio data takes time, making it unsuitable for large-scale data processing. Furthermore, conventional methods have the problem of making it difficult to set evaluation criteria that effectively utilize interview data and calculate the degree of match between candidates.

[1004] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1005] In this invention, the server includes means for starting recording at the start of the interview, means for stopping recording at the end of the interview and transmitting the audio data to the server, means for transcribing the audio data using speech recognition technology on the server, means for analyzing the transcribed data and evaluating logical thinking, line of sight, and emotions, means for calculating the degree of match with the company based on the analysis results, means for notifying the user of the results, and means for displaying the results. This enables the automation and optimization of the interview process, and enables efficient and fair evaluation.

[1006] The "means for starting recording at the start of an interview" is a function for enabling the built-in microphone of the terminal by user operation at the start of an interview and recording audio data.

[1007] "Means for stopping recording at the end of the interview and sending audio data to the server" is a function that stops recording at the end of the interview by user operation and sends the recorded audio data to the server using a secure protocol.

[1008] "Means for transcribing voice data using voice recognition technology on the server" is a function for converting voice data received by the server into text data using voice recognition technology.

[1009] "Means of analyzing transcribed data and evaluating logical thinking, perspective, and emotions" refers to a function that analyzes transcribed text data using a natural language processing model and sentiment analysis API to evaluate the test taker's logical thinking, perspective, and emotions.

[1010] "Means for calculating the degree of match with a company based on the analysis results" is a function for calculating the degree of compatibility between an applicant and a company from the results of data analysis, based on the evaluation criteria set by the company.

[1011] The "means for notifying the user of the results" is a function for sending the calculated analysis results and the degree of match to the user's terminal so that the user can check them.

[1012] The "means for displaying results" is a function for visually displaying the analysis results and the degree of match received on the terminal so that the user can easily check them.

[1013] An "automated conversational agent" is a conversational system such as a chatbot that uses AI, and is a technology for automating voice or text interactions with users.

[1014] "Huge interview data" refers to a collection of large amounts of recorded interview data, and is a variety of data related to interviews, including audio data, text data, video data, and the like.

[1015] A "training dataset" is a collection of data used to train and optimize a generative AI model.

[1016] The present invention provides a system that aims to improve the efficiency of the interview process and the accuracy of the evaluation. This system automates and optimizes the series of processes including recording, transmitting, transcribing, analyzing, calculating the match degree, and notifying the results of the interview. Specific embodiments and examples are described below.

[1017] System Overview

[1018] The system consists of a "terminal" operated by the "user" (interviewer or applicant) and a "server" that processes and analyzes voice data. The terminal and server are connected via a network, and various data is sent and received. The main hardware used includes PCs, tablets, and smartphones. The software used includes a voice recording application, a voice recognition engine, and a natural language processing model.

[1019] Program processing overview

[1020] 1. Start recording

[1021] When the user wants to start the interview, they press the "Start Recording" button on the dedicated application (e.g., "Interview Recorder app") installed on their device.

[1022] The device will activate its built-in microphone and record audio data.

[1023] 2. Sending audio data

[1024] After the interview is over, the user presses the "End Recording" button.

[1025] The device will stop recording and send the saved audio data to the server using a secure protocol (HTTPS).

[1026] 3. Transcription

[1027] The server transcribes the received audio data using the Google Cloud Speech-to-Text API.

[1028] The audio data is converted into text data.

[1029] 4. Data Analysis

[1030] The server analyzes the transcribed text data and evaluates the following:

[1031] An "NLP model" is used to evaluate logical thinking, specifically by analyzing the frequency of keyword occurrences and the logical structure of sentences.

[1032] To evaluate sentiment, we use the Sentiment Analysis API, which calculates positive and negative sentiment indicators from text.

[1033] If necessary, video data is also analyzed to analyze gaze and facial expressions using "OpenCV."

[1034] 5. Match Calculation

[1035] The server calculates the degree of match with the company from the analysis results based on the evaluation criteria set by the company.

[1036] The analysis results are input into a custom recommendation engine to calculate a match score.

[1037] 6. Result notification

[1038] The server transmits the analysis results and the degree of match to the terminal.

[1039] The terminal displays the results on the screen so that the user can check them.

[1040] Specific examples

[1041] Example 1: Conducting an in-person interview

[1042] When the user (interviewer) begins a face-to-face interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop recording. The device sends the recorded data to the server, which transcribes it using "Google Cloud Speech-to-Text." The NLP model then performs text analysis and sentiment analysis, and calculates the degree of match based on criteria set by the company. The analysis results and degree of match are sent to the device, where the user (interviewer) can check the results and download an evaluation report.

[1043] Example 2: Conducting an automated interview

[1044] The user (applicant) activates the automated interview function, and the AI ​​chatbot conducts the interview. The chatbot asks questions one by one, and the user (applicant) responds by voice or text. The server transcribes the responses in real time using Google Cloud Speech-to-Text and analyzes them using an NLP model. Finally, the degree of match is calculated, and the results are sent to the device. The user (applicant) can check the results.

[1045] In this way, by combining voice recognition technology with generative AI models, this system can significantly improve the efficiency and accuracy of interviews. It also enables multifaceted data analysis to select candidates who best fit the skill sets required by companies.

[1046] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1047] Step 1:

[1048] When the user starts the interview, he or she presses the "Start Recording" button on the dedicated application.

[1049] Input: User operation (instruction to start recording)

[1050] The device will activate the built-in microphone and begin recording.

[1051] Output: Recorded audio data

[1052] Step 2:

[1053] The terminal will display the recording status to let the user know that recording is in progress.

[1054] Input: Recording time and status information

[1055] The terminal displays the recording status on the user interface.

[1056] Output: Status display during recording

[1057] Step 3:

[1058] After the interview is over, the user presses the "End Recording" button.

[1059] Input: User operation (instruction to end recording)

[1060] The device will stop recording and temporarily save the audio data.

[1061] Output: Saved audio data

[1062] Step 4:

[1063] The terminal transmits the recorded voice data to the server.

[1064] Input: Saved audio data

[1065] The device sends the audio data to the server using the HTTPS protocol.

[1066] Output: Audio data sent to the server

[1067] Step 5:

[1068] The server receives the audio data and transcribes it using speech recognition technology.

[1069] Input: Received audio data

[1070] The server converts the audio into text data using the Google Cloud Speech-to-Text API.

[1071] Output: Transcribed text data

[1072] Step 6:

[1073] The server analyzes the transcribed text data and evaluates logical thinking, gaze, and emotions.

[1074] Input: Transcribed text data

[1075] The server does the following:

[1076] Evaluating logical thinking using an "NLP model" (analysis of keyword frequency and logical structure of sentences)

[1077] Evaluate sentiment using the Sentiment Analysis API (calculate positive and negative sentiment indicators from text)

[1078] Optionally, use OpenCV to analyze gaze and facial expressions from video data.

[1079] Output: Analysis results

[1080] Step 7:

[1081] The server calculates the degree of match with the company based on the analysis results.

[1082] Input: Analysis results and evaluation criteria set by the company

[1083] The server uses a custom recommendation engine to calculate the match score.

[1084] Output: Match score

[1085] Step 8:

[1086] The server transmits the analysis results and the degree of match to the terminal.

[1087] Input: Analysis results and match scores

[1088] The server sends the analysis results and the degree of match to the terminal.

[1089] Output: Submitted analysis results and match scores

[1090] Step 9:

[1091] The terminal displays the results on the screen so that the user can check them.

[1092] Input: Submitted analysis results and match rate

[1093] The terminal displays the results visually for easy confirmation by the user.

[1094] Output: Displayed analysis results and match scores

[1095] (Application example 1)

[1096] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1097] It is not easy for factory employees to efficiently learn new operations and procedures and improve their proficiency. Conventional methods require managers to manually evaluate each one, which is time-consuming and can lead to inequity. Furthermore, because worker aptitude assessments tend to be subjective, it is difficult to assign appropriate personnel.

[1098] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1099] In this invention, the server includes a means for recording voice data, a means for transmitting the recorded voice data to the server, a means for transcribing the voice data using voice recognition technology in the server, a means for analyzing the transcribed data and evaluating logical thinking, proficiency, and aptitude, a means for calculating the skill level of the worker based on the analysis results, and a means for notifying the user of the results. This automates the aptitude evaluation of workers in the factory, making it possible to improve the efficiency and fairness of the evaluation.

[1100] "Voice data" is digital information that records the voice and voice instructions of the worker.

[1101] "Recording" is the act of saving audio data and making it available for later use.

[1102] A "server" is a computer system that receives, analyzes, and stores voice data via a network.

[1103] "Speech recognition technology" is a technology that converts collected voice data into text.

[1104] "Transcription" is the process of converting audio data into text data.

[1105] "Data analysis" is a method of analyzing transcribed text data according to various criteria.

[1106] "Logical thinking" refers to the ability to think logically and is a skill for solving problems systematically and efficiently.

[1107] "Skill level" is an index that evaluates how skilled a worker is at a particular task or operation.

[1108] "Aptitude" is an indicator of how well a worker is suited to a particular role or task.

[1109] "Skill level" indicates the degree of technical ability and knowledge of a worker.

[1110] "Notification" refers to the act of informing the user of the analysis results and evaluations.

[1111] An "automated conversational agent" is a program that can interact with employees through voice and text.

[1112] A "training dataset" is a collection of data used to train a generative AI model.

[1113] "Network architecture" is the structural design that enables generative AI models to operate efficiently.

[1114] A "loss function" is a function used to minimize error when training a generative AI model.

[1115] MODE FOR CARRYING OUT THE INVENTION

[1116] The present invention provides a system for automating the aptitude evaluation and skill level determination of workers in a factory. Specific embodiments are described below.

[1117] System Overview

[1118] The system consists of a "terminal" operated by the worker and a "server" that processes and analyzes the voice data. The terminal and server are connected via a network, and various data is sent and received.

[1119] Program processing overview

[1120] The system has the following main functions:

[1121] 1. Audio data recording:

[1122] When a worker starts work, he or she presses the "Start Recording" button on the terminal, which then records the work voice and behavioral data.

[1123] 2. Sending audio data:

[1124] When the worker finishes the work and presses the "Stop Recording" button, the terminal stops recording and sends the recorded voice data to the server.

[1125] 3. Transcription:

[1126] The server transcribes the received voice data using voice recognition technology and converts it into text data.

[1127] 4. Data Analysis:

[1128] The server analyzes the converted text data and evaluates the worker's logical thinking, proficiency, and aptitude.

[1129] 5. Skill Level Calculation:

[1130] The server calculates the worker's skill level based on the evaluation results.

[1131] 6. Result notification:

[1132] The server transmits the evaluation result and skill level to the terminal, which displays them.

[1133] Hardware and software used

[1134] Hardware:

[1135] Terminal: Equipped with a microphone for recording audio data and buttons for starting and stopping recording.

[1136] Server: Equipped with a high-performance processor and large memory capacity, it can quickly analyze voice data.

[1137] software:

[1138] Speech recognition technology: Transcribing voice data using Google's speech recognition API.

[1139] Data Analysis: Use the Transformers library to analyze the sentiment, logical thinking, proficiency, and aptitude of the transcribed data.

[1140] Specific examples

[1141] For example, a worker in charge of explaining how to operate a new machine presses the start recording button on the terminal to begin the explanation. When the work is finished, the worker presses the stop recording button to stop recording the audio data, and the terminal sends the data to the server. The server transcribes the audio data and analyzes the text data to evaluate the worker's proficiency and aptitude, and sends the results to the terminal. This allows managers to check the evaluation results of the workers and provide appropriate guidance and placement.

[1142] Prompt Sentence Examples

[1143] Below are some example prompts that can be used as input to generative AI models:

[1144] "Transcribe the audio data, analyze the employee's logical thinking and proficiency, and calculate their aptitude."

[1145] This will allow for quick and efficient evaluation of workers in your factory.

[1146] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1147] Step 1:

[1148] Processing content: The user presses the "Start recording" button on the device.

[1149] Specific operation: Before starting work in the factory, the user presses the "Start Recording" button to enable the terminal's recording function.

[1150] Input and Output: The input is the user pressing a button, and the output is the device starting recording.

[1151] Step 2:

[1152] Processing details: The device records the work audio.

[1153] Specific operation: The terminal starts recording the worker's voice data using the microphone. The voice data being recorded is saved in a buffer in real time.

[1154] Input and output: The input is the worker's voice, and the voice data is stored in the terminal's buffer memory.

[1155] Step 3:

[1156] Processing content: The user presses the "Stop Recording" button on the device.

[1157] Specific operation: When the work is finished, the user presses the "End Recording" button on the device to stop recording.

[1158] Input and Output: Input is the user pressing a button, and output is the device stopping recording and saving the audio data to a file.

[1159] Step 4:

[1160] Processing content: The device sends voice data to the server.

[1161] Specific operation: The device sends the saved voice data to the server via the network. The data transfer uses the HTTP protocol.

[1162] Input and Output: The input is the audio data stored on the device, and the output is the audio file sent to the server.

[1163] Step 5:

[1164] Processing: The server transcribes the audio data.

[1165] Specific operation: The server converts the received voice data into text using speech recognition technology. It uses Google's speech recognition API.

[1166] Input and Output: The input is the audio data sent to the server, and the output is the transcribed text data.

[1167] Step 6:

[1168] Processing content: The server analyzes the text data.

[1169] Specific operation: The server uses the Transformers library to analyze the transcribed text data for sentiment, logical thinking, proficiency, and aptitude.

[1170] Input and output: The input is transcribed text data, and the output is the analytical results (emotions, logical thinking, proficiency, aptitude).

[1171] Step 7:

[1172] Processing content: The server calculates the skill level.

[1173] Specific operation: Based on the analysis results, the server calculates the worker's skill level. Calculating the skill level includes the process of comparing the analysis results with the evaluation criteria.

[1174] Inputs and Outputs: The inputs are the parameters of the analysis results, and the output is the calculated skill level.

[1175] Step 8:

[1176] Processing content: The server sends the evaluation results to the terminal.

[1177] Specific operation: The server sends the calculated skill level and evaluation results to the terminal. Data is transferred using the HTTP protocol.

[1178] Input and Output: The input is the skill level and the evaluation result, and the output is the evaluation result data sent to the terminal.

[1179] Step 9:

[1180] Processing content: The terminal displays the evaluation results to the user.

[1181] Specific operation: The terminal displays the received evaluation results on the user interface, and the user can view the evaluation results on the terminal screen.

[1182] Input and output: The input is the evaluation result data sent from the server, and the output is the evaluation result displayed on the terminal screen.

[1183] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1184] The present invention is a system for improving the efficiency and fairness of the interview process, and in particular incorporates an "emotion engine" that evaluates the user's emotions by comprehensively analyzing audio data and video data. Specific embodiments and their operation are described below.

[1185] System configuration

[1186] The system consists of a "terminal" that records and transmits voice data, a "server" that analyzes voice and video data, and an "emotion engine" that recognizes the user's emotions. These components work together while sending and receiving data via a network.

[1187] Program processing overview

[1188] 1. Start recording

[1189] The user clicks the Start Interview button. The device activates its audio and video recording functions and records audio and video data.

[1190] The device will display a confirmation message saying "Recording has started."

[1191] 2. Transmission of audio and video data

[1192] When the interview is finished, the user clicks the end button, and the device stops recording and saves the audio and video data.

[1193] The device sends the stored audio and video data to the server, which returns a confirmation message to the device, which displays it.

[1194] 3. Transcription and data analysis

[1195] The server passes the received voice data to a voice recognition engine to perform transcription.

[1196] The server passes the transcribed text data to an analysis engine, which evaluates logical thinking, analyzes emotions, and analyzes gaze.

[1197] 4. Operation of the Emotion Engine

[1198] The emotion engine of the server analyzes the user's emotion from the audio and video data.

[1199] Emotional data is analyzed in real time and incorporated into logical thinking and other assessments.

[1200] The server calculates the degree of match with the company, including the results of sentiment analysis.

[1201] 5. Notification of Results

[1202] The server sends the analysis results and the degree of match to the terminal, which then displays the results on the screen.

[1203] Users can review the results and download the report if desired.

[1204] Specific examples

[1205] Example 1: Conducting an in-person interview

[1206] When the user (interviewer or applicant) starts the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop the audio and video recording. The device sends the audio and video data to the server, which transcribes the audio data and analyzes emotions using an emotion engine. The evaluation results and match degree are calculated and sent to the device. The user (interviewer or applicant) can check the results and download the evaluation report.

[1207] Example 2: Conducting an automated interview

[1208] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes the voice in real time, analyzes emotions using an emotion engine, and analyzes logical thinking and eye contact. Finally, the analysis results are calculated along with the degree of match and sent to the device. The user (applicant) can check the results.

[1209] Emotion Engine Details

[1210] The emotion engine uses advanced machine learning algorithms to analyze a user's emotional state from audio and video data. Audio analysis uses features such as tone, speed, and pitch of the voice. Video analysis, on the other hand, evaluates facial expressions, eye contact, and other physical movements. The emotion engine integrates these data in real time to accurately recognize the user's emotional state.

[1211] Additional information

[1212] This system integrates voice recognition and emotion analysis technologies to improve the objectivity and fairness of interviews. Furthermore, it utilizes distributed computing technology on the server to efficiently process large amounts of interview data. This significantly improves the efficiency and accuracy of interviews, enabling companies to select more suitable candidates.

[1213] The above is a specific embodiment of the present invention. This system comprehensively analyzes audio data and video data to achieve highly accurate emotion evaluation, thereby contributing to the improvement of the interview process.

[1214] The processing flow will be explained below.

[1215] Step 1:

[1216] The user clicks the Start Interview button. The device activates the audio and video recording functions and begins recording audio and video data during the interview. The device displays a confirmation message on the screen saying "Recording has started."

[1217] Step 2:

[1218] When the interview is finished, the user clicks the end button, and the device stops recording audio and video, saving the recorded audio and video data.

[1219] Step 3:

[1220] The device sends the stored audio and video data to the server. The server receives the audio and video data and returns a confirmation message to the device stating, "Audio and video data has been successfully sent to the server." The device displays this confirmation message on its screen.

[1221] Step 4:

[1222] The server passes the received voice data to the voice recognition engine, which begins analyzing and transcribing the voice data. The voice recognition engine analyzes the voice data and converts it into text data. After conversion is complete, the server receives the text data.

[1223] Step 5:

[1224] The server then passes the video data to an analysis engine along with the transcribed text data, which then analyzes the user's gaze and facial expressions. The video analysis engine then analyzes the video data and extracts information about the user's gaze and facial expressions.

[1225] Step 6:

[1226] The server passes the text data, eye movement data, and facial expression data obtained from the analysis engine to the emotion engine, which analyzes the user's emotions using the tone, pitch, and speed of the voice and the facial expression information extracted from the video data.

[1227] Step 7:

[1228] The server's emotion engine synthesizes the analyzed emotion data and generates a rich emotion profile, which can be used as an indicator of the user's emotional state.

[1229] Step 8:

[1230] The server calculates the degree of match based on all analysis results, including the emotional profile, according to the evaluation criteria set by the company, which includes a multifaceted evaluation of logical thinking, eye contact, emotions, and voice content.

[1231] Step 9:

[1232] The server sends the analysis results and the degree of match to the device, which displays the results on the screen for the user to review. The user can then review the results and download a report if necessary.

[1233] Step 10:

[1234] (In the case of automated interviews)

[1235] The user selects "Automatic Interview" in the application. The device sends a request for an automated interview to the server. The server launches an automated conversation agent (AI chatbot) and asks the user the defined questions.

[1236] Step 11:

[1237] The AI ​​chatbot on the server asks the user questions one after another and receives answers via voice or text. The server records the exchange of questions and answers, performs real-time speech recognition and transcription, and also analyzes the user's emotions using an emotion engine.

[1238] Step 12:

[1239] The server evaluates all data obtained in real time and calculates the degree of match. Finally, the analysis results are sent to the device along with the degree of match. The user can check the results and download the report if necessary.

[1240] Example 2

[1241] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1242] The current interview process relies on a lot of manual work and subjective judgment, resulting in a lack of efficiency and fairness. In particular, because it relies on the interviewer's subjectivity, evaluation criteria are vague and emotions and logical thinking may not be properly evaluated. In addition, managing and analyzing large amounts of interview data is time-consuming, placing a heavy burden on companies. This often makes it difficult to select the right candidates.

[1243] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1244] In this invention, the server includes means for recording audio data and video data, means for transmitting the recorded audio data and video data to the server, means for transcribing the audio data using speech recognition technology in the server, means for analyzing the transcribed data and video data and evaluating logical thinking, line of sight, and emotions, means for calculating the degree of match with the company based on the analysis results, and means for notifying the user of the degree of match with the company and the analysis results, thereby making it possible to improve the efficiency and fairness of the interview process.

[1245] "Audio data" refers to voice information recorded during an interview or conversation that has been stored in digital form.

[1246] "Video data" refers to image and video information recorded during an interview or conversation that has been saved in digital format.

[1247] A "server" is a remote computer system used for data storage, analysis, and management.

[1248] "Voice recognition technology" is a technology for converting voice input into text data.

[1249] "Transcription" is the process of converting audio data into text data.

[1250] "Logical thinking" is an evaluation criterion that refers to the ability to think logically and reason.

[1251] "Gaze" refers to the direction and movement of the gaze, and refers to gaze behavior during an interview, etc.

[1252] "Emotions" refers to the psychological state judged from facial expressions, tone of voice, etc. during the interview.

[1253] "Evaluation" is the process of quantifying or grading the user's abilities and emotions based on the analyzed data.

[1254] "Match level" is an indicator of the compatibility between the interviewee and the company.

[1255] "User" refers to an interviewer or applicant who uses the system.

[1256] "Notification" refers to the act of transmitting information by the server to inform the user of the analysis results and evaluations.

[1257] The present invention provides a system for comprehensively analyzing audio data and video data to improve the efficiency and fairness of the interview process. Specific embodiments of this system will be described below.

[1258] System configuration

[1259] The system consists of three main components:

[1260] 1. "Devices" that record audio and video data

[1261] 2. The "server" that analyzes the received data and calculates the results

[1262] 3. "Emotion Engine" that analyzes emotional states

[1263] These components operate in cooperation with each other via a network.

[1264] Hardware and Software Use

[1265] Terminal

[1266] When the user starts the interview, the device will record audio and video data using the camera and microphone. The device also has audio and video recording functions to save the audio and video data. When the recording is finished, the data is sent to the server.

[1267] server

[1268] When the server receives the voice data, it transcribes it using the Google Cloud Speech-to-Text API. The resulting text data is then analyzed using IBM Watson Natural Language Understanding. The analysis includes logical thinking, eye gaze analysis, and sentiment analysis. Finally, the server calculates the match rate with the company based on the analysis results and sends the results to the device.

[1269] Emotion Engine

[1270] The server's emotion engine uses the Emotion API to analyze the user's emotional state from audio and video data. Audio analysis uses features such as tone, speed, and pitch of the voice. Video analysis evaluates facial expressions, eye movements, and other body movements.

[1271] Specific examples

[1272] Example 1: Conducting an in-person interview

[1273] When the user (interviewer or applicant) starts the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop the audio and video recording. The device sends the audio and video data to the server, which transcribes the audio data and analyzes emotions using an emotion engine. The evaluation results and match degree are calculated and sent to the device. The user (interviewer or applicant) can check the results and download the evaluation report.

[1274] Example 2: Conducting an automated interview

[1275] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes the voice in real time, analyzes emotions using an emotion engine, and analyzes logical thinking and eye contact. Finally, the analysis results are calculated along with the degree of match and sent to the device. The user (applicant) can check the results.

[1276] Prompt Sentence Examples

[1277] "Tell me about a system that can streamline and improve fairness in the interview process for applicants. How can you analyze emotions using audio and video data? Can you also give me some specific use cases?"

[1278] The above is an embodiment of the present invention. By using this system, it is possible to significantly improve the efficiency and fairness of interviews.

[1279] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1280] Step 1:

[1281] The user clicks the start interview button. The device responds by enabling the camera and microphone and starting to record audio and video data. The input here is the user's click action, and the output is the audio and video data that has started to be recorded. Specifically, pressing the record button initializes the device's camera and microphone.

[1282] Step 2:

[1283] The device records the audio and video data of the interview in real time. The recorded data is temporarily stored in the device's local storage. The input at this time is the user's voice and movements, and the output is the audio and video data stored in the local storage. Specifically, audio data is obtained from the microphone input, and video data is obtained from the camera input.

[1284] Step 3:

[1285] When the interview is over, the user clicks the end button. The device stops recording and saves the recorded data completely. The input at this time is the click of the end button, and the output is the final audio and video data file. Specifically, the recording is stopped and the data is saved.

[1286] Step 4:

[1287] The device sends the stored audio and video data to the server. The input here is the audio and video data stored in the device's local storage, and the output is the data sent to the server. Specifically, the data is uploaded to the server via the network using the HTTP or HTTPS protocol.

[1288] Step 5:

[1289] The server checks the received data and passes the audio data to the Google Cloud Speech-to-Text API for transcription. The input is the audio data received by the server, and the output is the transcribed text data. Specifically, the API is called to generate text data from the audio data.

[1290] Step 6:

[1291] The server passes the transcribed text data to IBM Watson Natural Language Understanding to evaluate logical thinking. The input here is the transcribed text data, and the output is the evaluation results. Specifically, the text data is analyzed and logical thinking ability is quantified.

[1292] Step 7:

[1293] At the same time, the server passes the video data to the Emotion API to analyze the emotional state. The input is the video data, and the output is the user's emotional state data. Specifically, the system analyzes facial expressions and gaze from the video data to recognize emotions.

[1294] Step 8:

[1295] The server integrates the logical thinking evaluation results and emotional state data to calculate the match degree with the company. The inputs are the evaluation results and emotional state data, and the output is a match degree score. Specifically, the server calculates the match degree by weighting each evaluation data.

[1296] Step 9:

[1297] The server sends the analysis results and match scores to the terminal. The input at this time is the analysis results and match scores at the server, and the output is notification data to the terminal. Specifically, the analysis results and evaluation scores are transferred to the terminal via the network.

[1298] Step 10:

[1299] The terminal displays the received analysis results and match scores on the screen. Finally, the user can check the results and download the evaluation report if necessary. The input in this case is the notification data from the server, and the output is the displayed analysis results and a downloadable evaluation report. Specifically, the user interface is updated and the results are displayed.

[1300] (Application example 2)

[1301] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1302] Conventional interview systems analyze only voice data, making it difficult to fully grasp the interviewer's emotions and intentions. Furthermore, there are insufficient means to improve the fairness and efficiency of interviews. It is also difficult to grasp customer emotions and provide appropriate service when serving customers in brick-and-mortar stores. Conventional technologies cannot analyze customer emotions in real time and suggest optimal customer service methods based on that analysis, making it difficult to improve customer satisfaction.

[1303] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1304] In this invention, the server includes a means for recording voice data, a means for transmitting the recorded voice data to the server, a means for transcribing the voice data using voice recognition technology on the server, a means for analyzing the transcribed data and evaluating logical thinking, gaze, and emotions, a means for calculating the degree of match with the company based on the analysis results, and a means for notifying the user of the results. This enables the interviewer's emotions to be recognized with high accuracy from voice and video data during the interview, enabling fair and efficient evaluation. Furthermore, in customer service at physical stores, customer satisfaction can be improved by analyzing customer emotions in real time using smart devices equipped with cameras and microphones and proposing optimal customer service methods based on the analysis results.

[1305] "Audio data" is a recording of sound collected through an audio input device such as a microphone.

[1306] A "server" is a computer system that transmits, receives, and processes data over a network.

[1307] "Voice recognition technology" is a technology that analyzes voice data and converts it into text format.

[1308] "Transcription" is the process of recognizing audio data and converting it into corresponding text data.

[1309] "Logical thinking" is the process of evaluating logical thinking.

[1310] "Gaze" is the process of assessing visual direction and focus.

[1311] "Emotion" is the process of assessing a user's mental state from audio and video data.

[1312] "Company match" is an indicator that evaluates the compatibility between the interviewer and the company.

[1313] "Notification" is the process of informing users of the analysis results.

[1314] A "camera" is a device that captures video data.

[1315] A "microphone" is a device that collects sound.

[1316] A "display device" is a device that visually presents analysis results and proposals.

[1317] The "Emotion Engine" is a system that analyzes emotions from audio and video data.

[1318] "Feedback" is the process of returning analysis results and suggestions to users.

[1319] An "automated conversational agent" is a system that uses artificial intelligence to automatically conduct dialogue.

[1320] A "smart device" is an electronic device with advanced features that can connect to the Internet.

[1321] A "generative AI model" is an artificial intelligence model that is trained using large amounts of data.

[1322] A "prompt" is a textual instruction that is input to a generative AI model.

[1323] The present invention provides a system for evaluating emotions by analyzing audio and video data during interviews and customer service, thereby realizing efficient and fair dialogue. Specific embodiments of the system are described below.

[1324] System configuration

[1325] The system consists of the following main components:

[1326] 1. Device: This includes smart devices equipped with cameras and microphones (e.g., smart glasses, smartphones, and head-mounted displays). These devices capture audio and video data in real time and have communication capabilities.

[1327] 2. Server: A cloud server for analyzing data, equipped with speech recognition technology, an emotion engine, and a data analysis engine. For example, Amazon Web Services (AWS) or Google Cloud Platform (GCP) are used.

[1328] 3. Emotion Engine: Software for analyzing emotions from audio and video data. Microsoft Azure Emotion API and Affectiva SDK are used.

[1329] 4. Generative AI model: An artificial intelligence model trained using large amounts of training data, which then generates appropriate feedback and suggestions.

[1330] Program processing

[1331] 1. Data capture and transmission:

[1332] A user (e.g., a store clerk) wears a terminal such as smart glasses and captures audio and video data.

[1333] The captured data is sent to a server in real time.

[1334] 2. Data Analysis:

[1335] The server transcribes the received voice data using a speech recognition engine, such as Google Cloud Speech-to-Text or IBM Watson Speech to Text.

[1336] Once transcribed, the text data is passed to an emotion engine that analyzes the emotions of the customer or interviewer. Voice analysis uses features such as tone, speed, and pitch, while video analysis evaluates facial expressions, eye movements, and other physical movements.

[1337] 3. Real-time feedback:

[1338] Based on the analysis results of the emotion engine, the server proposes optimal actions to the user (e.g., a store clerk) in real time, enabling the user to communicate efficiently with customers and interviewers.

[1339] The suggestions are fed back to the terminal's display device.

[1340] Specific examples

[1341] For example, imagine a salesperson wearing smart glasses explaining a luxury product to a customer. In this case, the glasses will analyze the customer's facial expressions and tone of voice in real time to gauge their level of interest or anxiety, and based on that, the glasses will suggest specific actions such as providing a more detailed explanation or offering a topic to help them relax.

[1342] Specific examples of prompts are as follows:

[1343] "Read the following sentence in a fully emotive tone: 'This watch is a classic design from the 1950s and is worn by many celebrities.'"

[1344] Using this prompt, the generative AI model can generate dialogue in specific scenarios and create data to improve the accuracy of its sentiment analysis.

[1345] The above is a specific embodiment of the present invention. This system realizes highly accurate emotion evaluation through integrated analysis of audio and video data, enabling efficient and fair operation in interviews and brick-and-mortar stores.

[1346] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1347] Step 1:

[1348] A user (e.g., a store clerk) puts on smart glasses and begins serving customers or conducting an interview. The smart glasses have a built-in camera and microphone to capture audio and video data. Pressing the device's activation button starts audio and video recording.

[1349] Input: User's audio and video

[1350] Output: Captured audio and video data

[1351] Step 2:

[1352] The device transmits the captured audio and video data to the server in real time, and during the transmission, the device temporarily stores the data to ensure network stability.

[1353] Input: Captured audio and video data

[1354] Output: Sending audio and video data to the server

[1355] Step 3:

[1356] The server then passes the received voice data to a speech recognition engine, such as Google Cloud Speech-to-Text or IBM Watson Speech to Text, which analyzes the voice data and generates corresponding text data.

[1357] Input: Audio data

[1358] Output: Transcribed text data

[1359] Step 4:

[1360] The server passes the transcribed text data and video data to the emotion engine for emotion analysis. The emotion engine extracts features from the audio and video data and analyzes emotions using a machine learning model.

[1361] Input: Transcribed text and video data

[1362] Output: Emotion analysis results

[1363] Step 5:

[1364] Based on the results of the sentiment analysis, the server generates the optimal action using a generative AI model, which combines past data with prompts to create specific suggestions for the user.

[1365] Input: Sentiment analysis results

[1366] Output: Recommendations for optimal action

[1367] Example prompt sentence:

[1368] "Read the following sentence in a fully emotive tone: 'This watch is a classic design from the 1950s and is worn by many celebrities.'"

[1369] Step 6:

[1370] The server sends the generated suggestions to the display device of the terminal and provides feedback to the user, who then sees the suggestions in real time on the display of the smart glasses.

[1371] Input: Generated proposals

[1372] Output: Feedback to a display device

[1373] Step 7:

[1374] The user takes appropriate action based on the displayed suggestions, which are then captured again by the camera and microphone and reflected in the next cycle.

[1375] Input: Shown suggestions

[1376] Output: User actions

[1377] The above are the specific steps of the processing of the program of the system that realizes the application example.

[1378] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1379] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1380] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1381] [Fourth embodiment]

[1382] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1383] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1384] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1385] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1386] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1387] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1388] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1389] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1390] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1391] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1392] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1393] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1394] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1395] The present invention provides a system that aims to improve the efficiency of the interview process and the accuracy of the evaluation. This system automates and optimizes the series of processes including recording, transmitting, transcribing, analyzing, calculating the match degree, and notifying the results of the interview. Specific embodiments and examples are described below.

[1396] System Overview

[1397] The system consists of a "terminal" operated by the "user" (interviewer or applicant) and a "server" that processes and analyzes the voice data. The terminal and server are connected via a network, and various data is sent and received.

[1398] Program processing overview

[1399] 1. Start recording

[1400] When the user (interviewer or applicant) starts the interview, he / she presses the "Start Recording" button on the terminal.

[1401] The device will have its recording function enabled and will record the audio data of the interview.

[1402] 2. Sending audio data

[1403] After the interview is over, the user presses the "End Recording" button on the terminal.

[1404] The device will stop recording and send the saved audio data to the server.

[1405] 3. Transcription

[1406] The server transcribes the received voice data using voice recognition technology.

[1407] The audio data is converted into text data.

[1408] 4. Data Analysis

[1409] The server analyzes the transcribed text data and evaluates the test-taker's logical thinking, gaze, and emotions.

[1410] If necessary, in the case of face-to-face interviews, video data will also be analyzed to analyze eye movements and facial expressions.

[1411] 5. Match Calculation

[1412] The server calculates the degree of match with the company from the analysis results based on the evaluation criteria set by the company.

[1413] The match level is an indicator of how closely the candidate matches the profile of the employee the company is looking for.

[1414] 6. Result notification

[1415] The server transmits the analysis results and the degree of match to the terminal.

[1416] The terminal displays the results on the screen so that the user can check them.

[1417] Specific examples

[1418] Example 1: Conducting an in-person interview

[1419] When the user (interviewer) begins the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop recording. The device sends the recorded data to the server, which transcribes the audio data and performs logical thinking and sentiment analysis. The match degree is calculated based on the analysis results and the results are sent to the device. The user (interviewer) can check the results and download an evaluation report.

[1420] Example 2: Conducting an automated interview

[1421] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes and analyzes the voice in real time. Finally, the analysis results calculate the degree of match based on the required criteria, and the results are sent to the device. The user (applicant) can check the results.

[1422] DETAILED DESCRIPTION OF THE EMBODIMENTS

[1423] The system includes multiple evaluation criteria to fairly evaluate candidates who best fit the skillsets desired by companies. By combining speech recognition technology with generative AI models, it achieves highly accurate transcription and sentiment analysis. Furthermore, the server's distributed computing technology allows for fast and efficient processing of even large data sets, significantly improving the efficiency of the entire interview process.

[1424] Additional information

[1425] In addition, the generative AI model is trained using a vast amount of interview data collected. The training process includes improvements to the network architecture and the introduction of a new loss function to optimize the model's learning speed and accuracy. Furthermore, multifaceted data analysis is incorporated to achieve a 360-degree evaluation of the company's desired talent profile.

[1426] The above is a specific embodiment of the present invention. This system not only significantly improves the efficiency and accuracy of interviews, but also eliminates interviewer bias and achieves fairer evaluations.

[1427] The processing flow will be explained below.

[1428] Step 1:

[1429] The user clicks the Start Interview button. The device activates the recording function and starts recording audio data. The device displays a confirmation message on the screen saying "Recording has started."

[1430] Step 2:

[1431] When the interview is finished, the user clicks the stop recording button. The device stops recording and saves the recorded audio data.

[1432] Step 3:

[1433] The device sends the saved voice data to the server. The server receives the voice data and returns a confirmation message to the device stating, "The voice data has been successfully sent to the server." The device displays this confirmation message on its screen.

[1434] Step 4:

[1435] The server passes the received voice data to the voice recognition engine, which then analyzes and transcribes the voice data. The voice recognition engine analyzes the voice data and converts it into text data.

[1436] Step 5:

[1437] The server passes the transcribed text data to the analysis engine, which then uses the text data to evaluate logical thinking, analyze emotions, and analyze gaze.

[1438] Step 6:

[1439] (For face-to-face interviews)

[1440] The device sends the video data collected during recording to a server, which then passes the video data to an analysis engine for analysis of gaze and facial expressions.

[1441] Step 7:

[1442] The server calculates the degree of match based on the data obtained by the analysis engine and in accordance with the evaluation criteria set by the company.

[1443] Step 8:

[1444] The server sends the match rate and analysis results to the terminal, which displays these results on the screen so that the user can check them.

[1445] Step 9:

[1446] (In the case of automated interviews)

[1447] The user selects "Automatic Interview" in the application. The device sends a request for an automated interview to the server. The server launches an automated conversation agent (AI chatbot) and asks questions to the user.

[1448] Step 10:

[1449] The AI ​​chatbot on the server asks the user predefined questions in sequence and receives answers in voice or text. The server records the exchange of questions and answers and performs real-time speech recognition, transcription, and analysis.

[1450] Step 11:

[1451] The server calculates the analysis results and the degree of match and sends them to the terminal. The user can check the results and download the report if necessary.

[1452] Example 1

[1453] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1454] The current interview process requires a lot of manual work, is inefficient, and the interviewer's subjectivity has a significant impact on the evaluation, making it difficult to achieve fair and accurate evaluation. Furthermore, transcribing and analyzing audio data takes time, making it unsuitable for large-scale data processing. Furthermore, conventional methods have the problem of making it difficult to set evaluation criteria that effectively utilize interview data and calculate the degree of match between candidates.

[1455] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1456] In this invention, the server includes means for starting recording at the start of the interview, means for stopping recording at the end of the interview and transmitting the audio data to the server, means for transcribing the audio data using speech recognition technology on the server, means for analyzing the transcribed data and evaluating logical thinking, line of sight, and emotions, means for calculating the degree of match with the company based on the analysis results, means for notifying the user of the results, and means for displaying the results. This enables the automation and optimization of the interview process, and enables efficient and fair evaluation.

[1457] The "means for starting recording at the start of an interview" is a function for enabling the built-in microphone of the terminal by user operation at the start of an interview and recording audio data.

[1458] "Means for stopping recording at the end of the interview and sending audio data to the server" is a function that stops recording at the end of the interview by user operation and sends the recorded audio data to the server using a secure protocol.

[1459] "Means for transcribing voice data using voice recognition technology on the server" is a function for converting voice data received by the server into text data using voice recognition technology.

[1460] "Means of analyzing transcribed data and evaluating logical thinking, perspective, and emotions" refers to a function that analyzes transcribed text data using a natural language processing model and sentiment analysis API to evaluate the test taker's logical thinking, perspective, and emotions.

[1461] "Means for calculating the degree of match with a company based on the analysis results" is a function for calculating the degree of compatibility between an applicant and a company from the results of data analysis, based on the evaluation criteria set by the company.

[1462] The "means for notifying the user of the results" is a function for sending the calculated analysis results and the degree of match to the user's terminal so that the user can check them.

[1463] The "means for displaying results" is a function for visually displaying the analysis results and the degree of match received on the terminal so that the user can easily check them.

[1464] An "automated conversational agent" is a conversational system such as a chatbot that uses AI, and is a technology for automating voice or text interactions with users.

[1465] "Huge interview data" refers to a collection of large amounts of recorded interview data, and is a variety of data related to interviews, including audio data, text data, video data, and the like.

[1466] A "training dataset" is a collection of data used to train and optimize a generative AI model.

[1467] The present invention provides a system that aims to improve the efficiency of the interview process and the accuracy of the evaluation. This system automates and optimizes the series of processes including recording, transmitting, transcribing, analyzing, calculating the match degree, and notifying the results of the interview. Specific embodiments and examples are described below.

[1468] System Overview

[1469] The system consists of a "terminal" operated by the "user" (interviewer or applicant) and a "server" that processes and analyzes voice data. The terminal and server are connected via a network, and various data is sent and received. The main hardware used includes PCs, tablets, and smartphones. The software used includes a voice recording application, a voice recognition engine, and a natural language processing model.

[1470] Program processing overview

[1471] 1. Start recording

[1472] When the user wants to start the interview, they press the "Start Recording" button on the dedicated application (e.g., "Interview Recorder app") installed on their device.

[1473] The device will activate its built-in microphone and record audio data.

[1474] 2. Sending audio data

[1475] After the interview is over, the user presses the "End Recording" button.

[1476] The device will stop recording and send the saved audio data to the server using a secure protocol (HTTPS).

[1477] 3. Transcription

[1478] The server transcribes the received audio data using the Google Cloud Speech-to-Text API.

[1479] The audio data is converted into text data.

[1480] 4. Data Analysis

[1481] The server analyzes the transcribed text data and evaluates the following:

[1482] An "NLP model" is used to evaluate logical thinking, specifically by analyzing the frequency of keyword occurrences and the logical structure of sentences.

[1483] To evaluate sentiment, we use the Sentiment Analysis API, which calculates positive and negative sentiment indicators from text.

[1484] If necessary, video data is also analyzed to analyze gaze and facial expressions using "OpenCV."

[1485] 5. Match Calculation

[1486] The server calculates the degree of match with the company from the analysis results based on the evaluation criteria set by the company.

[1487] The analysis results are input into a custom recommendation engine to calculate a match score.

[1488] 6. Result notification

[1489] The server transmits the analysis results and the degree of match to the terminal.

[1490] The terminal displays the results on the screen so that the user can check them.

[1491] Specific examples

[1492] Example 1: Conducting an in-person interview

[1493] When the user (interviewer) begins a face-to-face interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop recording. The device sends the recorded data to the server, which transcribes it using "Google Cloud Speech-to-Text." The NLP model then performs text analysis and sentiment analysis, and calculates the degree of match based on criteria set by the company. The analysis results and degree of match are sent to the device, where the user (interviewer) can check the results and download an evaluation report.

[1494] Example 2: Conducting an automated interview

[1495] The user (applicant) activates the automated interview function, and the AI ​​chatbot conducts the interview. The chatbot asks questions one by one, and the user (applicant) responds by voice or text. The server transcribes the responses in real time using Google Cloud Speech-to-Text and analyzes them using an NLP model. Finally, the degree of match is calculated, and the results are sent to the device. The user (applicant) can check the results.

[1496] In this way, by combining voice recognition technology with generative AI models, this system can significantly improve the efficiency and accuracy of interviews. It also enables multifaceted data analysis to select candidates who best fit the skill sets required by companies.

[1497] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1498] Step 1:

[1499] When the user starts the interview, he or she presses the "Start Recording" button on the dedicated application.

[1500] Input: User operation (instruction to start recording)

[1501] The device will activate the built-in microphone and begin recording.

[1502] Output: Recorded audio data

[1503] Step 2:

[1504] The terminal will display the recording status to let the user know that recording is in progress.

[1505] Input: Recording time and status information

[1506] The terminal displays the recording status on the user interface.

[1507] Output: Status display during recording

[1508] Step 3:

[1509] After the interview is over, the user presses the "End Recording" button.

[1510] Input: User operation (instruction to end recording)

[1511] The device will stop recording and temporarily save the audio data.

[1512] Output: Saved audio data

[1513] Step 4:

[1514] The terminal transmits the recorded voice data to the server.

[1515] Input: Saved audio data

[1516] The device sends the audio data to the server using the HTTPS protocol.

[1517] Output: Audio data sent to the server

[1518] Step 5:

[1519] The server receives the audio data and transcribes it using speech recognition technology.

[1520] Input: Received audio data

[1521] The server converts the audio into text data using the Google Cloud Speech-to-Text API.

[1522] Output: Transcribed text data

[1523] Step 6:

[1524] The server analyzes the transcribed text data and evaluates logical thinking, gaze, and emotions.

[1525] Input: Transcribed text data

[1526] The server does the following:

[1527] Evaluating logical thinking using an "NLP model" (analysis of keyword frequency and logical structure of sentences)

[1528] Evaluate sentiment using the Sentiment Analysis API (calculate positive and negative sentiment indicators from text)

[1529] Optionally, use OpenCV to analyze gaze and facial expressions from video data.

[1530] Output: Analysis results

[1531] Step 7:

[1532] The server calculates the degree of match with the company based on the analysis results.

[1533] Input: Analysis results and evaluation criteria set by the company

[1534] The server uses a custom recommendation engine to calculate the match score.

[1535] Output: Match score

[1536] Step 8:

[1537] The server transmits the analysis results and the degree of match to the terminal.

[1538] Input: Analysis results and match scores

[1539] The server sends the analysis results and the degree of match to the terminal.

[1540] Output: Submitted analysis results and match scores

[1541] Step 9:

[1542] The terminal displays the results on the screen so that the user can check them.

[1543] Input: Submitted analysis results and match rate

[1544] The terminal displays the results visually for easy confirmation by the user.

[1545] Output: Displayed analysis results and match scores

[1546] (Application example 1)

[1547] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1548] It is not easy for factory employees to efficiently learn new operations and procedures and improve their proficiency. Conventional methods require managers to manually evaluate each one, which is time-consuming and can lead to inequity. Furthermore, because worker aptitude assessments tend to be subjective, it is difficult to assign appropriate personnel.

[1549] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1550] In this invention, the server includes a means for recording voice data, a means for transmitting the recorded voice data to the server, a means for transcribing the voice data using voice recognition technology in the server, a means for analyzing the transcribed data and evaluating logical thinking, proficiency, and aptitude, a means for calculating the skill level of the worker based on the analysis results, and a means for notifying the user of the results. This automates the aptitude evaluation of workers in the factory, making it possible to improve the efficiency and fairness of the evaluation.

[1551] "Voice data" is digital information that records the voice and voice instructions of the worker.

[1552] "Recording" is the act of saving audio data and making it available for later use.

[1553] A "server" is a computer system that receives, analyzes, and stores voice data via a network.

[1554] "Speech recognition technology" is a technology that converts collected voice data into text.

[1555] "Transcription" is the process of converting audio data into text data.

[1556] "Data analysis" is a method of analyzing transcribed text data according to various criteria.

[1557] "Logical thinking" refers to the ability to think logically and is a skill for solving problems systematically and efficiently.

[1558] "Skill level" is an index that evaluates how skilled a worker is at a particular task or operation.

[1559] "Aptitude" is an indicator of how well a worker is suited to a particular role or task.

[1560] "Skill level" indicates the degree of technical ability and knowledge of a worker.

[1561] "Notification" refers to the act of informing the user of the analysis results and evaluations.

[1562] An "automated conversational agent" is a program that can interact with employees through voice and text.

[1563] A "training dataset" is a collection of data used to train a generative AI model.

[1564] "Network architecture" is the structural design that enables generative AI models to operate efficiently.

[1565] A "loss function" is a function used to minimize error when training a generative AI model.

[1566] MODE FOR CARRYING OUT THE INVENTION

[1567] The present invention provides a system for automating the aptitude evaluation and skill level determination of workers in a factory. Specific embodiments are described below.

[1568] System Overview

[1569] The system consists of a "terminal" operated by the worker and a "server" that processes and analyzes the voice data. The terminal and server are connected via a network, and various data is sent and received.

[1570] Program processing overview

[1571] The system has the following main functions:

[1572] 1. Audio data recording:

[1573] When a worker starts work, he or she presses the "Start Recording" button on the terminal, which then records the work voice and behavioral data.

[1574] 2. Sending audio data:

[1575] When the worker finishes the work and presses the "Stop Recording" button, the terminal stops recording and sends the recorded voice data to the server.

[1576] 3. Transcription:

[1577] The server transcribes the received voice data using voice recognition technology and converts it into text data.

[1578] 4. Data Analysis:

[1579] The server analyzes the converted text data and evaluates the worker's logical thinking, proficiency, and aptitude.

[1580] 5. Skill Level Calculation:

[1581] The server calculates the worker's skill level based on the evaluation results.

[1582] 6. Result notification:

[1583] The server transmits the evaluation result and skill level to the terminal, which displays them.

[1584] Hardware and software used

[1585] Hardware:

[1586] Terminal: Equipped with a microphone for recording audio data and buttons for starting and stopping recording.

[1587] Server: Equipped with a high-performance processor and large memory capacity, it can quickly analyze voice data.

[1588] software:

[1589] Speech recognition technology: Transcribing voice data using Google's speech recognition API.

[1590] Data Analysis: Use the Transformers library to analyze the sentiment, logical thinking, proficiency, and aptitude of the transcribed data.

[1591] Specific examples

[1592] For example, a worker in charge of explaining how to operate a new machine presses the start recording button on the terminal to begin the explanation. When the work is finished, the worker presses the stop recording button to stop recording the audio data, and the terminal sends the data to the server. The server transcribes the audio data and analyzes the text data to evaluate the worker's proficiency and aptitude, and sends the results to the terminal. This allows managers to check the evaluation results of the workers and provide appropriate guidance and placement.

[1593] Prompt Sentence Examples

[1594] Below are some example prompts that can be used as input to generative AI models:

[1595] "Transcribe the audio data, analyze the employee's logical thinking and proficiency, and calculate their aptitude."

[1596] This will allow for quick and efficient evaluation of workers in your factory.

[1597] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1598] Step 1:

[1599] Processing content: The user presses the "Start recording" button on the device.

[1600] Specific operation: Before starting work in the factory, the user presses the "Start Recording" button to enable the terminal's recording function.

[1601] Input and Output: The input is the user pressing a button, and the output is the device starting recording.

[1602] Step 2:

[1603] Processing details: The device records the work audio.

[1604] Specific operation: The terminal starts recording the worker's voice data using the microphone. The voice data being recorded is saved in a buffer in real time.

[1605] Input and output: The input is the worker's voice, and the voice data is stored in the terminal's buffer memory.

[1606] Step 3:

[1607] Processing content: The user presses the "Stop Recording" button on the device.

[1608] Specific operation: When the work is finished, the user presses the "End Recording" button on the device to stop recording.

[1609] Input and Output: Input is the user pressing a button, and output is the device stopping recording and saving the audio data to a file.

[1610] Step 4:

[1611] Processing content: The device sends voice data to the server.

[1612] Specific operation: The device sends the saved voice data to the server via the network. The data transfer uses the HTTP protocol.

[1613] Input and Output: The input is the audio data stored on the device, and the output is the audio file sent to the server.

[1614] Step 5:

[1615] Processing: The server transcribes the audio data.

[1616] Specific operation: The server converts the received voice data into text using speech recognition technology. It uses Google's speech recognition API.

[1617] Input and Output: The input is the audio data sent to the server, and the output is the transcribed text data.

[1618] Step 6:

[1619] Processing content: The server analyzes the text data.

[1620] Specific operation: The server uses the Transformers library to analyze the transcribed text data for sentiment, logical thinking, proficiency, and aptitude.

[1621] Input and output: The input is transcribed text data, and the output is the analytical results (emotions, logical thinking, proficiency, aptitude).

[1622] Step 7:

[1623] Processing content: The server calculates the skill level.

[1624] Specific operation: Based on the analysis results, the server calculates the worker's skill level. Calculating the skill level includes the process of comparing the analysis results with the evaluation criteria.

[1625] Inputs and Outputs: The inputs are the parameters of the analysis results, and the output is the calculated skill level.

[1626] Step 8:

[1627] Processing content: The server sends the evaluation results to the terminal.

[1628] Specific operation: The server sends the calculated skill level and evaluation results to the terminal. Data is transferred using the HTTP protocol.

[1629] Input and Output: The input is the skill level and the evaluation result, and the output is the evaluation result data sent to the terminal.

[1630] Step 9:

[1631] Processing content: The terminal displays the evaluation results to the user.

[1632] Specific operation: The terminal displays the received evaluation results on the user interface, and the user can view the evaluation results on the terminal screen.

[1633] Input and output: The input is the evaluation result data sent from the server, and the output is the evaluation result displayed on the terminal screen.

[1634] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1635] The present invention is a system for improving the efficiency and fairness of the interview process, and in particular incorporates an "emotion engine" that evaluates the user's emotions by comprehensively analyzing audio data and video data. Specific embodiments and their operation are described below.

[1636] System configuration

[1637] The system consists of a "terminal" that records and transmits voice data, a "server" that analyzes voice and video data, and an "emotion engine" that recognizes the user's emotions. These components work together while sending and receiving data via a network.

[1638] Program processing overview

[1639] 1. Start recording

[1640] The user clicks the Start Interview button. The device activates its audio and video recording functions and records audio and video data.

[1641] The device will display a confirmation message saying "Recording has started."

[1642] 2. Transmission of audio and video data

[1643] When the interview is finished, the user clicks the end button, and the device stops recording and saves the audio and video data.

[1644] The device sends the stored audio and video data to the server, which returns a confirmation message to the device, which displays it.

[1645] 3. Transcription and data analysis

[1646] The server passes the received voice data to a voice recognition engine to perform transcription.

[1647] The server passes the transcribed text data to an analysis engine, which evaluates logical thinking, analyzes emotions, and analyzes gaze.

[1648] 4. Operation of the Emotion Engine

[1649] The emotion engine of the server analyzes the user's emotion from the audio and video data.

[1650] Emotional data is analyzed in real time and incorporated into logical thinking and other assessments.

[1651] The server calculates the degree of match with the company, including the results of sentiment analysis.

[1652] 5. Notification of Results

[1653] The server sends the analysis results and the degree of match to the terminal, which then displays the results on the screen.

[1654] Users can review the results and download the report if desired.

[1655] Specific examples

[1656] Example 1: Conducting an in-person interview

[1657] When the user (interviewer or applicant) starts the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop the audio and video recording. The device sends the audio and video data to the server, which transcribes the audio data and analyzes emotions using an emotion engine. The evaluation results and match degree are calculated and sent to the device. The user (interviewer or applicant) can check the results and download the evaluation report.

[1658] Example 2: Conducting an automated interview

[1659] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes the voice in real time, analyzes emotions using an emotion engine, and analyzes logical thinking and eye contact. Finally, the analysis results are calculated along with the degree of match and sent to the device. The user (applicant) can check the results.

[1660] Emotion Engine Details

[1661] The emotion engine uses advanced machine learning algorithms to analyze a user's emotional state from audio and video data. Audio analysis uses features such as tone, speed, and pitch of the voice. Video analysis, on the other hand, evaluates facial expressions, eye contact, and other physical movements. The emotion engine integrates these data in real time to accurately recognize the user's emotional state.

[1662] Additional information

[1663] This system integrates voice recognition and emotion analysis technologies to improve the objectivity and fairness of interviews. Furthermore, it utilizes distributed computing technology on the server to efficiently process large amounts of interview data. This significantly improves the efficiency and accuracy of interviews, enabling companies to select more suitable candidates.

[1664] The above is a specific embodiment of the present invention. This system comprehensively analyzes audio data and video data to achieve highly accurate emotion evaluation, thereby contributing to the improvement of the interview process.

[1665] The processing flow will be explained below.

[1666] Step 1:

[1667] The user clicks the Start Interview button. The device activates the audio and video recording functions and begins recording audio and video data during the interview. The device displays a confirmation message on the screen saying "Recording has started."

[1668] Step 2:

[1669] When the interview is finished, the user clicks the end button, and the device stops recording audio and video, saving the recorded audio and video data.

[1670] Step 3:

[1671] The device sends the stored audio and video data to the server. The server receives the audio and video data and returns a confirmation message to the device stating, "Audio and video data has been successfully sent to the server." The device displays this confirmation message on its screen.

[1672] Step 4:

[1673] The server passes the received voice data to the voice recognition engine, which begins analyzing and transcribing the voice data. The voice recognition engine analyzes the voice data and converts it into text data. After conversion is complete, the server receives the text data.

[1674] Step 5:

[1675] The server then passes the video data to an analysis engine along with the transcribed text data, which then analyzes the user's gaze and facial expressions. The video analysis engine then analyzes the video data and extracts information about the user's gaze and facial expressions.

[1676] Step 6:

[1677] The server passes the text data, eye movement data, and facial expression data obtained from the analysis engine to the emotion engine, which analyzes the user's emotions using the tone, pitch, and speed of the voice and the facial expression information extracted from the video data.

[1678] Step 7:

[1679] The server's emotion engine synthesizes the analyzed emotion data and generates a rich emotion profile, which can be used as an indicator of the user's emotional state.

[1680] Step 8:

[1681] The server calculates the degree of match based on all analysis results, including the emotional profile, according to the evaluation criteria set by the company, which includes a multifaceted evaluation of logical thinking, eye contact, emotions, and voice content.

[1682] Step 9:

[1683] The server sends the analysis results and the degree of match to the device, which displays the results on the screen for the user to review. The user can then review the results and download a report if necessary.

[1684] Step 10:

[1685] (In the case of automated interviews)

[1686] The user selects "Automatic Interview" in the application. The device sends a request for an automated interview to the server. The server launches an automated conversation agent (AI chatbot) and asks the user the defined questions.

[1687] Step 11:

[1688] The AI ​​chatbot on the server asks the user questions one after another and receives answers via voice or text. The server records the exchange of questions and answers, performs real-time speech recognition and transcription, and also analyzes the user's emotions using an emotion engine.

[1689] Step 12:

[1690] The server evaluates all data obtained in real time and calculates the degree of match. Finally, the analysis results are sent to the device along with the degree of match. The user can check the results and download the report if necessary.

[1691] Example 2

[1692] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1693] The current interview process relies on a lot of manual work and subjective judgment, resulting in a lack of efficiency and fairness. In particular, because it relies on the interviewer's subjectivity, evaluation criteria are vague and emotions and logical thinking may not be properly evaluated. In addition, managing and analyzing large amounts of interview data is time-consuming, placing a heavy burden on companies. This often makes it difficult to select the right candidates.

[1694] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1695] In this invention, the server includes means for recording audio data and video data, means for transmitting the recorded audio data and video data to the server, means for transcribing the audio data using speech recognition technology in the server, means for analyzing the transcribed data and video data and evaluating logical thinking, line of sight, and emotions, means for calculating the degree of match with the company based on the analysis results, and means for notifying the user of the degree of match with the company and the analysis results, thereby making it possible to improve the efficiency and fairness of the interview process.

[1696] "Audio data" refers to voice information recorded during an interview or conversation that has been stored in digital form.

[1697] "Video data" refers to image and video information recorded during an interview or conversation that has been saved in digital format.

[1698] A "server" is a remote computer system used for data storage, analysis, and management.

[1699] "Voice recognition technology" is a technology for converting voice input into text data.

[1700] "Transcription" is the process of converting audio data into text data.

[1701] "Logical thinking" is an evaluation criterion that refers to the ability to think logically and reason.

[1702] "Gaze" refers to the direction and movement of the gaze, and refers to gaze behavior during an interview, etc.

[1703] "Emotions" refers to the psychological state judged from facial expressions, tone of voice, etc. during the interview.

[1704] "Evaluation" is the process of quantifying or grading the user's abilities and emotions based on the analyzed data.

[1705] "Match level" is an indicator of the compatibility between the interviewee and the company.

[1706] "User" refers to an interviewer or applicant who uses the system.

[1707] "Notification" refers to the act of transmitting information by the server to inform the user of the analysis results and evaluations.

[1708] The present invention provides a system for comprehensively analyzing audio data and video data to improve the efficiency and fairness of the interview process. Specific embodiments of this system will be described below.

[1709] System configuration

[1710] The system consists of three main components:

[1711] 1. "Devices" that record audio and video data

[1712] 2. The "server" that analyzes the received data and calculates the results

[1713] 3. "Emotion Engine" that analyzes emotional states

[1714] These components operate in cooperation with each other via a network.

[1715] Hardware and Software Use

[1716] Terminal

[1717] When the user starts the interview, the device will record audio and video data using the camera and microphone. The device also has audio and video recording functions to save the audio and video data. When the recording is finished, the data is sent to the server.

[1718] server

[1719] When the server receives the voice data, it transcribes it using the Google Cloud Speech-to-Text API. The resulting text data is then analyzed using IBM Watson Natural Language Understanding. The analysis includes logical thinking, eye gaze analysis, and sentiment analysis. Finally, the server calculates the match rate with the company based on the analysis results and sends the results to the device.

[1720] Emotion Engine

[1721] The server's emotion engine uses the Emotion API to analyze the user's emotional state from audio and video data. Audio analysis uses features such as tone, speed, and pitch of the voice. Video analysis evaluates facial expressions, eye movements, and other body movements.

[1722] Specific examples

[1723] Example 1: Conducting an in-person interview

[1724] When the user (interviewer or applicant) starts the interview, they press the "Start Recording" button on the dedicated application. When the interview is finished, they press the "Stop Recording" button to stop the audio and video recording. The device sends the audio and video data to the server, which transcribes the audio data and analyzes emotions using an emotion engine. The evaluation results and match degree are calculated and sent to the device. The user (interviewer or applicant) can check the results and download the evaluation report.

[1725] Example 2: Conducting an automated interview

[1726] The user (applicant) activates the automated interview function, and the AI ​​chatbot begins the automated interview. The chatbot asks questions in sequence, and the applicant responds via voice or text. The server transcribes the voice in real time, analyzes emotions using an emotion engine, and analyzes logical thinking and eye contact. Finally, the analysis results are calculated along with the degree of match and sent to the device. The user (applicant) can check the results.

[1727] Prompt Sentence Examples

[1728] "Tell me about a system that can streamline and improve fairness in the interview process for applicants. How can you analyze emotions using audio and video data? Can you also give me some specific use cases?"

[1729] The above is an embodiment of the present invention. By using this system, it is possible to significantly improve the efficiency and fairness of interviews.

[1730] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1731] Step 1:

[1732] The user clicks the start interview button. The device responds by enabling the camera and microphone and starting to record audio and video data. The input here is the user's click action, and the output is the audio and video data that has started to be recorded. Specifically, pressing the record button initializes the device's camera and microphone.

[1733] Step 2:

[1734] The device records the audio and video data of the interview in real time. The recorded data is temporarily stored in the device's local storage. The input at this time is the user's voice and movements, and the output is the audio and video data stored in the local storage. Specifically, audio data is obtained from the microphone input, and video data is obtained from the camera input.

[1735] Step 3:

[1736] When the interview is over, the user clicks the end button. The device stops recording and saves the recorded data completely. The input at this time is the click of the end button, and the output is the final audio and video data file. Specifically, the recording is stopped and the data is saved.

[1737] Step 4:

[1738] The device sends the stored audio and video data to the server. The input here is the audio and video data stored in the device's local storage, and the output is the data sent to the server. Specifically, the data is uploaded to the server via the network using the HTTP or HTTPS protocol.

[1739] Step 5:

[1740] The server checks the received data and passes the audio data to the Google Cloud Speech-to-Text API for transcription. The input is the audio data received by the server, and the output is the transcribed text data. Specifically, the API is called to generate text data from the audio data.

[1741] Step 6:

[1742] The server passes the transcribed text data to IBM Watson Natural Language Understanding to evaluate logical thinking. The input here is the transcribed text data, and the output is the evaluation results. Specifically, the text data is analyzed and logical thinking ability is quantified.

[1743] Step 7:

[1744] At the same time, the server passes the video data to the Emotion API to analyze the emotional state. The input is the video data, and the output is the user's emotional state data. Specifically, the system analyzes facial expressions and gaze from the video data to recognize emotions.

[1745] Step 8:

[1746] The server integrates the logical thinking evaluation results and emotional state data to calculate the match degree with the company. The inputs are the evaluation results and emotional state data, and the output is a match degree score. Specifically, the server calculates the match degree by weighting each evaluation data.

[1747] Step 9:

[1748] The server sends the analysis results and match scores to the terminal. The input at this time is the analysis results and match scores at the server, and the output is notification data to the terminal. Specifically, the analysis results and evaluation scores are transferred to the terminal via the network.

[1749] Step 10:

[1750] The terminal displays the received analysis results and match scores on the screen. Finally, the user can check the results and download the evaluation report if necessary. The input in this case is the notification data from the server, and the output is the displayed analysis results and a downloadable evaluation report. Specifically, the user interface is updated and the results are displayed.

[1751] (Application example 2)

[1752] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1753] Conventional interview systems analyze only voice data, making it difficult to fully grasp the interviewer's emotions and intentions. Furthermore, there are insufficient means to improve the fairness and efficiency of interviews. It is also difficult to grasp customer emotions and provide appropriate service when serving customers in brick-and-mortar stores. Conventional technologies cannot analyze customer emotions in real time and suggest optimal customer service methods based on that analysis, making it difficult to improve customer satisfaction.

[1754] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1755] In this invention, the server includes a means for recording voice data, a means for transmitting the recorded voice data to the server, a means for transcribing the voice data using voice recognition technology on the server, a means for analyzing the transcribed data and evaluating logical thinking, gaze, and emotions, a means for calculating the degree of match with the company based on the analysis results, and a means for notifying the user of the results. This enables the interviewer's emotions to be recognized with high accuracy from voice and video data during the interview, enabling fair and efficient evaluation. Furthermore, in customer service at physical stores, customer satisfaction can be improved by analyzing customer emotions in real time using smart devices equipped with cameras and microphones and proposing optimal customer service methods based on the analysis results.

[1756] "Audio data" is a recording of sound collected through an audio input device such as a microphone.

[1757] A "server" is a computer system that transmits, receives, and processes data over a network.

[1758] "Voice recognition technology" is a technology that analyzes voice data and converts it into text format.

[1759] "Transcription" is the process of recognizing audio data and converting it into corresponding text data.

[1760] "Logical thinking" is the process of evaluating logical thinking.

[1761] "Gaze" is the process of assessing visual direction and focus.

[1762] "Emotion" is the process of assessing a user's mental state from audio and video data.

[1763] "Company match" is an indicator that evaluates the compatibility between the interviewer and the company.

[1764] "Notification" is the process of informing users of the analysis results.

[1765] A "camera" is a device that captures video data.

[1766] A "microphone" is a device that collects sound.

[1767] A "display device" is a device that visually presents analysis results and proposals.

[1768] The "Emotion Engine" is a system that analyzes emotions from audio and video data.

[1769] "Feedback" is the process of returning analysis results and suggestions to users.

[1770] An "automated conversational agent" is a system that uses artificial intelligence to automatically conduct dialogue.

[1771] A "smart device" is an electronic device with advanced features that can connect to the Internet.

[1772] A "generative AI model" is an artificial intelligence model that is trained using large amounts of data.

[1773] A "prompt" is a textual instruction that is input to a generative AI model.

[1774] The present invention provides a system for evaluating emotions by analyzing audio and video data during interviews and customer service, thereby realizing efficient and fair dialogue. Specific embodiments of the system are described below.

[1775] System configuration

[1776] The system consists of the following main components:

[1777] 1. Device: This includes smart devices equipped with cameras and microphones (e.g., smart glasses, smartphones, and head-mounted displays). These devices capture audio and video data in real time and have communication capabilities.

[1778] 2. Server: A cloud server for analyzing data, equipped with speech recognition technology, an emotion engine, and a data analysis engine. For example, Amazon Web Services (AWS) or Google Cloud Platform (GCP) are used.

[1779] 3. Emotion Engine: Software for analyzing emotions from audio and video data. Microsoft Azure Emotion API and Affectiva SDK are used.

[1780] 4. Generative AI model: An artificial intelligence model trained using large amounts of training data, which then generates appropriate feedback and suggestions.

[1781] Program processing

[1782] 1. Data capture and transmission:

[1783] A user (e.g., a store clerk) wears a terminal such as smart glasses and captures audio and video data.

[1784] The captured data is sent to a server in real time.

[1785] 2. Data Analysis:

[1786] The server transcribes the received voice data using a speech recognition engine, such as Google Cloud Speech-to-Text or IBM Watson Speech to Text.

[1787] Once transcribed, the text data is passed to an emotion engine that analyzes the emotions of the customer or interviewer. Voice analysis uses features such as tone, speed, and pitch, while video analysis evaluates facial expressions, eye movements, and other physical movements.

[1788] 3. Real-time feedback:

[1789] Based on the analysis results of the emotion engine, the server proposes optimal actions to the user (e.g., a store clerk) in real time, enabling the user to communicate efficiently with customers and interviewers.

[1790] The suggestions are fed back to the terminal's display device.

[1791] Specific examples

[1792] For example, imagine a salesperson wearing smart glasses explaining a luxury product to a customer. In this case, the glasses will analyze the customer's facial expressions and tone of voice in real time to gauge their level of interest or anxiety, and based on that, the glasses will suggest specific actions such as providing a more detailed explanation or offering a topic to help them relax.

[1793] Specific examples of prompts are as follows:

[1794] "Read the following sentence in a fully emotive tone: 'This watch is a classic design from the 1950s and is worn by many celebrities.'"

[1795] Using this prompt, the generative AI model can generate dialogue in specific scenarios and create data to improve the accuracy of its sentiment analysis.

[1796] The above is a specific embodiment of the present invention. This system realizes highly accurate emotion evaluation through integrated analysis of audio and video data, enabling efficient and fair operation in interviews and brick-and-mortar stores.

[1797] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1798] Step 1:

[1799] A user (e.g., a store clerk) puts on smart glasses and begins serving customers or conducting an interview. The smart glasses have a built-in camera and microphone to capture audio and video data. Pressing the device's activation button starts audio and video recording.

[1800] Input: User's audio and video

[1801] Output: Captured audio and video data

[1802] Step 2:

[1803] The device transmits the captured audio and video data to the server in real time, and during the transmission, the device temporarily stores the data to ensure network stability.

[1804] Input: Captured audio and video data

[1805] Output: Sending audio and video data to the server

[1806] Step 3:

[1807] The server then passes the received voice data to a speech recognition engine, such as Google Cloud Speech-to-Text or IBM Watson Speech to Text, which analyzes the voice data and generates corresponding text data.

[1808] Input: Audio data

[1809] Output: Transcribed text data

[1810] Step 4:

[1811] The server passes the transcribed text data and video data to the emotion engine for emotion analysis. The emotion engine extracts features from the audio and video data and analyzes emotions using a machine learning model.

[1812] Input: Transcribed text and video data

[1813] Output: Emotion analysis results

[1814] Step 5:

[1815] Based on the results of the sentiment analysis, the server generates the optimal action using a generative AI model, which combines past data with prompts to create specific suggestions for the user.

[1816] Input: Sentiment analysis results

[1817] Output: Recommendations for optimal action

[1818] Example prompt sentence:

[1819] "Read the following sentence in a fully emotive tone: 'This watch is a classic design from the 1950s and is worn by many celebrities.'"

[1820] Step 6:

[1821] The server sends the generated suggestions to the display device of the terminal and provides feedback to the user, who then sees the suggestions in real time on the display of the smart glasses.

[1822] Input: Generated proposals

[1823] Output: Feedback to a display device

[1824] Step 7:

[1825] The user takes appropriate action based on the displayed suggestions, which are then captured again by the camera and microphone and reflected in the next cycle.

[1826] Input: Shown suggestions

[1827] Output: User actions

[1828] The above are the specific steps of the processing of the program of the system that realizes the application example.

[1829] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1830] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1831] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1832] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1833] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1834] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1835] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1836] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1837] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1838] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1839] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1840] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1841] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1842] 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.

[1843] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1844] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1845] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1846] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1847] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1848] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1849] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1850] The following is further disclosed regarding the above embodiment.

[1851] (Claim 1)

[1852] means for recording audio data;

[1853] means for transmitting the recorded voice data to a server;

[1854] A means for transcribing the voice data using voice recognition technology on a server;

[1855] Analyzing the transcribed data to evaluate logical thinking, perspective, and emotions,

[1856] A means for calculating the degree of match with a company based on the analysis results;

[1857] a means for notifying the user of the results;

[1858] A system including:

[1859] (Claim 2)

[1860] a means for using an automated conversational agent in analyzing the speech data;

[1861] a means for the automated conversational agent to ask questions of the applicant and collect voice data;

[1862] A means to transcribe and analyze question and answer exchanges in real time,

[1863] 10. The system of claim 1, comprising:

[1864] (Claim 3)

[1865] A means for the server to collect and manage large amounts of interview data;

[1866] A means to efficiently transform the collected data into a training dataset;

[1867] A means to train generative AI models by improving network architectures and introducing new loss functions;

[1868] 10. The system of claim 1, comprising:

[1869] (Claim 4)

[1870] A means for the server to perform a 360-degree evaluation and calculate the match level based on the criteria set by the company;

[1871] A means to notify company personnel of the analysis results in real time,

[1872] 10. The system of claim 1, comprising:

[1873] "Example 1"

[1874] (Claim 1)

[1875] A means to start recording at the beginning of the interview;

[1876] means for stopping the recording at the end of the interview and transmitting the audio data to a server;

[1877] A means for transcribing the voice data using voice recognition technology on a server;

[1878] Analyzing the transcribed data to evaluate logical thinking, perspective, and emotions,

[1879] A means for calculating the degree of match with a company based on the analysis results;

[1880] a means for notifying the user of the results;

[1881] a means for displaying the results;

[1882] A system including:

[1883] (Claim 2)

[1884] a means for using an automated conversational agent in analyzing the speech data;

[1885] a means for the automated conversational agent to ask questions of the applicant and collect voice data;

[1886] A means to transcribe and analyze question and answer exchanges in real time,

[1887] 10. The system of claim 1, comprising:

[1888] (Claim 3)

[1889] A means for the server to collect and manage large amounts of interview data;

[1890] A means to efficiently transform the collected data into a training dataset;

[1891] A means to train generative AI models by improving network architectures and introducing new loss functions;

[1892] 10. The system of claim 1, comprising:

[1893] "Application Example 1"

[1894] (Claim 1)

[1895] means for recording audio data;

[1896] means for transmitting the recorded voice data to a server;

[1897] A means for transcribing the voice data using voice recognition technology on a server;

[1898] A means of analyzing the transcribed data and assessing logical thinking, proficiency, and aptitude;

[1899] a means for calculating a skill level of a worker based on the analysis result;

[1900] a means for notifying the user of the results;

[1901] A system including:

[1902] (Claim 2)

[1903] a means for using an automated conversational agent in analyzing the speech data;

[1904] means for the automated conversational agent to ask questions of the employee and collect speech data;

[1905] A means to transcribe and analyze question and answer exchanges in real time,

[1906] 10. The system of claim 1, comprising:

[1907] (Claim 3)

[1908] A means for the server to collect and manage a huge amount of operation instruction data;

[1909] A means to efficiently transform the collected data into a training dataset;

[1910] A means to train generative AI models by improving network architectures and introducing new loss functions;

[1911] 10. The system of claim 1, comprising:

[1912] "Example 2: Combining Emotion Engines"

[1913] (Claim 1)

[1914] means for recording audio and video data;

[1915] means for transmitting the recorded audio and video data to a server;

[1916] A means for transcribing the voice data using voice recognition technology on a server;

[1917] A means of analyzing transcribed data and video data to evaluate logical thinking, gaze, and emotions;

[1918] A means for calculating the degree of match with a company based on the analysis results;

[1919] A means for notifying users of the degree of match with companies and the analysis results;

[1920] A system including:

[1921] (Claim 2)

[1922] means for using an automated conversational agent in analyzing the audio and video data;

[1923] means for the automated conversational agent to ask questions of the applicant and collect audio and video data;

[1924] a means for transcribing and analyzing question and answer exchanges in real time;

[1925] 10. The system of claim 1, comprising:

[1926] (Claim 3)

[1927] A means for the server to collect and manage large amounts of interview data;

[1928] A means to efficiently transform the collected data into a training dataset;

[1929] A means to train generative AI models by improving network architectures and introducing new loss functions;

[1930] A means for feeding back the processing results to the user;

[1931] 10. The system of claim 1, comprising:

[1932] "Application example 2 when combining emotion engines"

[1933] (Claim 1)

[1934] means for recording audio data;

[1935] means for transmitting the recorded voice data to a server;

[1936] A means for transcribing the voice data using voice recognition technology on a server;

[1937] Analyzing the transcribed data to evaluate logical thinking, perspective, and emotions,

[1938] A means for calculating the degree of match with a company based on the analysis results;

[1939] a means for notifying the user of the results;

[1940] means for capturing audio and video data of a user in real time using a display device equipped with a camera and a microphone;

[1941] A means of suggesting appropriate actions to users in real time based on the results of analyzing audio and video data using an emotion engine;

[1942] a means for providing feedback to the user on the actions they take based on the suggested actions;

[1943] A system including:

[1944] (Claim 2)

[1945] a means for using an automated conversational agent in analyzing the speech data;

[1946] a means for the automated conversational agent to ask questions of the applicant and collect voice data;

[1947] A means to transcribe and analyze question and answer exchanges in real time,

[1948] Furthermore, in customer service situations, smart devices equipped with cameras and microphones can be used to analyze customer emotions and provide feedback in real time.

[1949] 10. The system of claim 1, comprising:

[1950] (Claim 3)

[1951] A means for the server to collect and manage large amounts of interview data;

[1952] A means to efficiently transform the collected data into a training dataset;

[1953] A means to train generative AI models by improving network architectures and introducing new loss functions;

[1954] further comprising: a means for using the results of the sentiment analysis of the audio and video data to train a generative AI model;

[1955] 10. The system of claim 1, comprising: [Explanation of symbols]

[1956] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. means for recording audio data; means for transmitting the recorded voice data to a server; A means for transcribing the voice data using voice recognition technology on a server; Analyzing the transcribed data to evaluate logical thinking, perspective, and emotions, A means for calculating the degree of match with a company based on the analysis results; a means for notifying the user of the results; A system including:

2. a means for using an automated conversational agent in analyzing the speech data; a means for the automated conversational agent to ask questions of the applicant and collect voice data; A means to transcribe and analyze question and answer exchanges in real time, The system of claim 1 , comprising:

3. A means for the server to collect and manage large amounts of interview data; A means to efficiently transform the collected data into a training dataset; A means to train generative AI models by improving network architectures and introducing new loss functions; The system of claim 1 , comprising:

4. A means for the server to perform a 360-degree evaluation and calculate the match level based on the criteria set by the company; A means to notify company personnel of the analysis results in real time, The system of claim 1 , comprising:

Images