System

Abstract

A system is provided.SOLUTION: A system including means for collecting voice data and motion data received by a care worker wearing a wearable device, means for analyzing the collected voice data and determining the presence or absence of abuse, means for analyzing the collected motion data and determining the presence or absence of violence, means for quantifying the degree of abuse or violence based on the analysis result, means for adjusting a reward of the care worker based on the quantified information, and means for linking the quantified information with a job information system.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] The high turnover rate among care workers is mainly due to stress in the workplace and dissatisfaction with compensation. The current situation, where verbal and physical abuse from clients is particularly common and appropriate measures are not being taken, is problematic. Furthermore, there is no system in place to visualize the stress experienced by care workers and reflect it in compensation, which can easily lead to mismatches when choosing a workplace. Given this background, it is necessary to quantify the stress experienced by care workers and adjust compensation accordingly in order to reduce turnover rates. [Means for solving the problem]

[0005] The present invention provides a means for collecting voice data and movement data received by caregivers wearing wearable devices, a means for analyzing the collected voice data to determine whether abusive language has been used, a means for analyzing the collected movement data to determine whether violence has been used, a means for quantifying the degree of abusive language and violence based on the analysis results, a means for adjusting the remuneration of caregivers based on the quantified information, and a means for linking the quantified information to a job information system.

[0006] This makes it possible to visualize the stress experienced by care workers and appropriately reflect it in their compensation. Furthermore, by including this quantified stress information in job postings, job seekers can understand the work environment in advance, preventing mismatches before and after starting work. These measures can reduce stress among care workers and eliminate dissatisfaction with compensation, thereby reducing turnover and supporting the retention and development of human resources.

[0007] A "wearable device" is a device that can be worn on the body, has sensors and communication functions, and collects, records, and transmits various data in real time.

[0008] "Audio data" refers to information that records sound waveforms as digital signals collected using a microphone.

[0009] "Motion data" refers to information recorded as digital signals of body movements using an acceleration sensor or the like.

[0010] "Abusive language" refers to words intended to hurt others or insulting remarks, and is a type of linguistic expression that can be identified through analysis of audio data.

[0011] "Violence" refers to physical actions that cause or are intended to cause physical harm to another person, and is identified through analysis of motion data.

[0012] "Analysis" refers to the process of analyzing collected voice and movement data and extracting specific patterns and information.

[0013] "Quantification" refers to expressing the degree of verbal abuse or violence numerically based on specific criteria, based on the results of analysis.

[0014] "Remuneration adjustment" refers to changing the amount of compensation, such as salary or bonus, depending on the degree of verbal abuse or violence that care workers are subjected to.

[0015] A "job information system" is a system that provides job information and makes it publicly available for job seekers to view.

[0016] A "stress score" is a numerical indicator of the stress that care workers experience while on the job, calculated based on the frequency and intensity of verbal and physical abuse. [Brief explanation of the drawings]

[0017] [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

[0018] 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.

[0019] First, the terms used in the following description will be explained.

[0020] 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).

[0021] 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.

[0022] 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.

[0023] 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.

[0024] 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."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 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.

[0028] 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).

[0029] 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.

[0030] 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.

[0031] 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.

[0032] 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.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 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.

[0035] 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.

[0036] 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.

[0037] 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."

[0038] This invention is a system for visualizing stress in nursing care settings and reflecting it in compensation. This system is implemented through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, and providing information.

[0039] Program processing

[0040] 1. Data Collection

[0041] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using a microphone, and movement data is measured using an accelerometer. This data is periodically transferred to a server.

[0042] 2. Data Analysis

[0043] The server analyzes the received voice data using a voice recognition algorithm to identify the presence and frequency of abusive language, and uses a behavior analysis algorithm to detect violent behavior from the movement data. This increases the detection rate of abusive language and violence, and allows for more accurate data to be obtained.

[0044] 3. Quantification

[0045] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. This score is calculated based on a set of criteria and can be aggregated weekly, monthly, or yearly.

[0046] 4. Remuneration adjustment

[0047] The server automatically adjusts the compensation of each care worker based on the numerical stress score. Specifically, it calculates the amount of bonuses and allowances according to the score and reflects this in the payroll system.

[0048] 5. Information provision

[0049] The server then links the quantified stress scores to a job information system and makes them available to job seekers, allowing them to understand the working environment in advance and choose an appropriate workplace.

[0050] Specific examples

[0051] For example, let's say that caregiver Sato is wearing a wearable device and encounters the following situation.

[0052] Monday

[0053] While working, Sato was subjected to verbal abuse from users, such as "idiot" and "useless," twice. The accelerometer also recorded him being hit in the chest once.

[0054] Tuesday

[0055] Sato was subjected to one verbal abuse, "Stop it!", but no violence occurred.

[0056] The server receives this data and uses voice recognition and behavioral analysis algorithms to determine two cases of verbal abuse and one case of violence, then quantifies the results and reflects them in the weekly score.

[0057] Based on the stress score obtained in this way, the server automatically adjusts Sato's compensation and reflects it in the payroll system. At the same time, this quantified information is also linked to the job information system, allowing other job seekers to view and understand it in real time.

[0058] Through the above process, the present invention aims to reduce turnover by visualizing the stress of care workers and appropriately reflecting this in their compensation. This system will improve the working environment for care workers and support the recruitment and development of human resources throughout the care industry.

[0059] The processing flow will be explained below.

[0060] Step 1:

[0061] The user, a caregiver, puts on the wearable device and turns it on, and the device is automatically paired with the server.

[0062] Step 2:

[0063] The device collects real-time audio data through the wearable device's microphone, while also collecting body movement data using an accelerometer.

[0064] Step 3:

[0065] The device stores the collected voice data and motion data in a buffer at regular intervals (for example, every minute). The stored data is given a timestamp.

[0066] Step 4:

[0067] The terminal transmits all the data in the buffer to the server at once. If a transmission error occurs, the terminal performs a retransmission process.

[0068] Step 5:

[0069] The server performs pre-processing of the received audio data, which includes noise filtering and voice normalization.

[0070] Step 6:

[0071] The server runs a speech recognition algorithm to convert the preprocessed audio data into text, from which it uses natural language processing (NLP) to extract abusive keywords.

[0072] Step 7:

[0073] The server pre-processes the motion data, filtering the accelerometer data to remove inaccurate data.

[0074] Step 8:

[0075] The server runs a behavioral analysis algorithm to detect patterns of violence in the filtered motion data, based on sudden movements and impacts.

[0076] Step 9:

[0077] The server calculates a stress score for each caregiver based on the extracted abusive keywords and the detected patterns of violence. The score is calculated based on certain criteria.

[0078] Step 10:

[0079] The server aggregates weekly, monthly, and yearly stress scores and adjusts compensation accordingly, which is then linked to the payroll system.

[0080] Step 11:

[0081] The server connects the calculated stress score to the job information system and makes it publicly available for job seekers to view.

[0082] Step 12:

[0083] Users can browse job information pages and check the stress scores of each workplace to choose a workplace that suits them.

[0084] Example 1

[0085] 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."

[0086] In the field of nursing care, caregivers often experience high levels of stress on a daily basis, and are frequently subjected to verbal and physical abuse. This leads to a decline in staff motivation and a high turnover rate. However, there is no way to objectively grasp the actual level of stress felt by caregivers, and appropriate compensation adjustments are not being made. To solve this issue, there is a need for a system that can quantify the stress of caregivers and automatically adjust compensation based on this information.

[0087] 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.

[0088] In this invention, the server includes means for collecting voice data and movement data received by caregivers wearing wearable devices, means for analyzing the collected voice data and determining whether abusive language has been used, means for analyzing the collected movement data and determining whether violence has been used, means for quantifying the degree of abusive language and violence based on the analysis results, means for automatically adjusting the caregiver's compensation based on the quantified information, means for linking the quantified information to a recruitment system and providing it, means for collecting voice data using an acoustic sensor and movement data using a motion sensor, and means for aggregating the quantified information on a weekly, monthly, and yearly basis and automatically adjusting compensation based on the quantified information. This makes it possible to objectively quantify the stress of caregivers and adjust compensation appropriately.

[0089] A "wearable device" is an electronic device that is worn on the body and is used to collect voice data and movement data.

[0090] "Voice data" refers to the audio information received by caregivers, and is digital data collected using microphones and acoustic sensors.

[0091] "Motion data" refers to information that measures the physical movements and acceleration of care workers, and is digital data collected using acceleration sensors and motion sensors.

[0092] "Analysis" is the process of using collected data to extract specific information and make decisions based on that information.

[0093] "Abusive language" refers to insulting words or expressions directed at nursing care staff.

[0094] "Violence" refers to physical aggression or force used against care staff.

[0095] "Quantification" is the process of expressing analyzed data as quantitative numbers.

[0096] "Automatic compensation adjustment" is a system that automatically calculates and adjusts the salaries and bonuses of care workers based on quantified information.

[0097] The "recruitment system" refers to an online platform where job seekers can view information about nursing care workplaces and apply.

[0098] An "acoustic sensor" is a device that detects sound information and collects it as digital data.

[0099] A "motion sensor" is a device that detects body movements and collects them as digital data.

[0100] This invention is a system for visualizing stress in nursing care settings and reflecting it in compensation. This system is implemented through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, and providing information.

[0101] 1. Data Collection

[0102] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using an acoustic sensor (microphone), and movement data is measured using a motion sensor (accelerometer). This data is periodically transferred to a server.

[0103] 2. Data Transfer

[0104] The collected voice and movement data is transferred to a server in real time via Wi-Fi or Bluetooth, for example, every minute, split into packets and sent with error checking to prevent data loss.

[0105] 3. Data Analysis

[0106] The server converts the received audio data into text using the Google Cloud Speech-to-Text API and uses natural language processing (NLP) algorithms to identify the presence and frequency of abusive language. The movement data is analyzed using a custom behavioral analysis model using TensorFlow to detect physical violence. For example, certain keywords or high accelerations can be detected to identify abusive language or violence.

[0107] 4. Quantification

[0108] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. The quantification criteria are 10 points for each verbal abuse incident and 50 points for each physical abuse incident, and the data can be compiled weekly, monthly, or yearly.

[0109] 5. Remuneration adjustment

[0110] The server automatically adjusts the compensation of each care worker based on the quantified stress score. Specifically, it calculates the amount of bonuses and allowances according to the stress score and reflects this in the payroll system (e.g., SAP SuccessFactors). For example, it can be set up so that an additional bonus is paid to staff whose monthly score exceeds 300 points.

[0111] 6. Information provision

[0112] The server connects the quantified stress score to recruitment systems (e.g., Recruit and Indeed) so that job seekers can understand the working environment in advance. For example, if Mr. Sato has a high stress score, that information will be displayed on the recruitment information page, allowing other job seekers to use it as a reference when choosing a workplace.

[0113] Specific examples

[0114] For example, let's say that caregiver Sato is wearing a wearable device and encounters the following situation.

[0115] Monday

[0116] While working, Sato was twice subjected to abusive language from users, such as being called an idiot and being useless, and an accelerometer recorded him being hit in the chest once.

[0117] Tuesday

[0118] Sato was subjected to one verbal abuse of "Stop it!" but no violence occurred.

[0119] The server receives this data and, using a custom behavioral analysis model using the Google Cloud Speech-to-Text API and TensorFlow, determines two instances of verbal abuse and one instance of violence, converts the results into a numerical value, and reflects this in a weekly score. Based on the stress score obtained in this way, the server automatically adjusts Mr. Sato's compensation and reflects it in the payroll system (e.g., SAP SuccessFactors). At the same time, this numerical information is also linked to job information systems (e.g., Recruit and Indeed), allowing other job seekers to view and understand it in real time.

[0120] Example prompts to input to the generative AI model

[0121] Prompt: Using the data below, explain the specific steps for a system that calculates stress scores for caregivers and automatically adjusts compensation based on the results.

[0122] Monday: Sato was subjected to verbal abuse such as "idiot" and "useless" twice during work hours, and an accelerometer recorded him being hit in the chest once.

[0123] Tuesday: One verbal abuse of "Stop it!", no violence occurred.

[0124] The system proceeds as follows:

[0125] 1. Data collection (collection of voice data and movement data, transfer to server)

[0126] 2. Data analysis (voice recognition and behavior analysis)

[0127] 3. Quantification (calculating stress scores)

[0128] 4. Compensation adjustment (automatic compensation adjustment based on numerical data)

[0129] 5. Information provision (linking to recruitment information systems)

[0130] This invention aims to reduce turnover by visualizing stress in the nursing care field and appropriately reflecting it in compensation. This system will improve the working environment for nursing care workers and support the recruitment and development of human resources throughout the nursing care industry.

[0131] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0132] Step 1: Data collection

[0133] The terminal starts up the wearable device worn by the caregiver and starts collecting voice and movement data. Voice data is collected using an acoustic sensor, and movement data is measured using a motion sensor. For example, voice data (such as "idiot" or "useless") is collected here, and the acceleration sensor detects the action of hitting the chest. The input of this step is the voice and movement of the caregiver, and the output is the collected raw data.

[0134] Step 2: Data Transfer

[0135] The device transfers the collected voice and movement data to the server in real time via Wi-Fi or Bluetooth. For example, the voice and movement data is compressed every minute and sent to the server in packet format, minimizing data loss. The input of this step is the collected raw data, and the output is the raw data stored on the server.

[0136] Step 3: Data analysis

[0137] The server converts the received voice data into text using the Google Cloud Speech-to-Text API and analyzes the presence and frequency of abusive language using natural language processing (NLP) algorithms. It also analyzes the behavioral data using a custom behavioral analysis model using TensorFlow to identify violent behavior. For example, it detects keywords (e.g., "idiot" or "stop") or high accelerations in the converted text. The input of this step is the received raw data, and the output is the analysis results.

[0138] Step 4: Quantify

[0139] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. For example, points are assigned, such as 10 points for each verbal abuse and 50 points for each physical abuse, and the totals are used to calculate the stress score. The input for this step is the analysis results, and the output is the calculated stress score.

[0140] Step 5: Compensation Adjustment

[0141] The server automatically adjusts the compensation of each caregiver based on the quantified stress score. Specifically, it calculates the amount of bonuses and allowances according to the stress score and connects this to the payroll system. For example, if the monthly score exceeds 300 points, a bonus is added. The input to this step is the calculated stress score, and the output is the adjusted compensation information.

[0142] Step 6: Provide information

[0143] The server links the quantified stress score to the job information system and provides the data so that job seekers can view it. For example, a job seeker can check the stress environment of Mr. Sato's workplace on the job information page. The input of this step is the adjusted compensation information, and the output is the stress score reflected in the job information system.

[0144] Through these steps, this system will improve the working environment in nursing care settings, appropriately reflect the stress of nursing care staff in their compensation, and enable job seekers to understand the working environment in advance.

[0145] (Application example 1)

[0146] 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."

[0147] Currently, in many brick-and-mortar stores, there are many situations where staff members who serve customers feel stressed. When this stress increases, employee motivation decreases and it may even lead to an increase in turnover. Furthermore, in order to improve customer satisfaction, it is important to create an environment where staff can work comfortably, but there is a lack of means to achieve this. Furthermore, there is a need to introduce an appropriate compensation system to make staff stress visible and improve the working environment.

[0148] 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.

[0149] In this invention, the server includes means for collecting voice data and movement data received by workers wearing wearable devices, means for analyzing the collected voice data and determining whether or not there are words that cause stress, means for analyzing the collected movement data and determining whether or not there are movement events that indicate stress, means for quantifying the stress level based on the analysis results, means for adjusting the worker's remuneration based on the quantified information, means for linking the quantified information to a recruitment information system, and means for providing the quantified information to the worker. This makes it possible to grasp staff stress in real time, adjust remuneration based on that information, and provide a comfortable working environment.

[0150] A "wearable device" is a device that can be worn on the body and collects voice data and movement data in real time.

[0151] "Employee" refers to a worker who provides customer service in a physical store.

[0152] "Audio data" refers to audio signals collected using a microphone and recorded as digital data.

[0153] "Motion data" is digital data indicating body movements obtained using an acceleration sensor or other sensor device.

[0154] "Stressor words" refer to specific words or phrases that are important when assessing an employee's stress level, and are often negative comments from customers or words that indicate overwork.

[0155] "Movement event" refers to a specific physical movement detected using sensors, and may include movements that are indicative of stress.

[0156] The "stress level" is the estimated level of stress felt by the worker, calculated from the collected voice data and movement data.

[0157] "Remuneration" refers to the monetary compensation, including salary and bonuses, that an employee receives for their work.

[0158] A "job information system" is an online system that handles information on the labor market and provides job seekers with data on companies and jobs.

[0159] "Means of providing" refers to methods or systems for communicating collected data and calculated results to employees, such as notification systems or portal sites.

[0160] MODE FOR CARRYING OUT THE INVENTION

[0161] To realize this invention, store employees who serve customers must wear a wearable device. The wearable device is used to collect voice and motion data in real time. The collected data is sent to a server and analyzed using voice recognition and motion analysis algorithms.

[0162] A description of what the program does

[0163] Hardware and software configuration:

[0164] Hardware:

[0165] Wearable device: Equipped with a microphone and accelerometer to collect voice and movement data.

[0166] Accelerometer: A sensor that detects body movement.

[0167] Microphone: A device for collecting sound.

[0168] software:

[0169] Server: A computer system that performs analysis and stores data.

[0170] Speech recognition algorithms: convert collected voice data into text and detect specific stressful words.

[0171] Motion analysis algorithm: Analyzes acceleration data to detect motion events that indicate stress. As a concrete example, peak detection is performed using the Python library "scipy."

[0172] Data processing and calculations:

[0173] The voice data is converted into text by a speech recognition algorithm on the server, which checks whether certain keywords (e.g., "busy," "tired," "terrible," etc.) are included to determine the stress factor.

[0174] Movement data is analyzed based on the output of the acceleration sensor to detect movement events that are indicators of stress (e.g., large or unnatural movements).

[0175] These analysis results are quantified and the employee's stress score is calculated.

[0176] Information provision and compensation adjustment:

[0177] The quantified information is stored on a server and compiled periodically, weekly, monthly, and annually, and employee compensation is automatically adjusted based on the results.

[0178] This information will also be linked to the job information system so that job seekers can view it.

[0179] The collected and analyzed stress information is also provided to employees, helping them understand their own stress levels and consider ways to improve them.

[0180] Examples:

[0181] For example, if a customer service staff member at a brick-and-mortar store hears a comment from a customer that they are "busy" or "tired" and begins to feel stressed during their shift, the voice data is collected through a microphone. An accelerometer is used to simultaneously detect movement events indicative of stress, such as hand vibrations. This data is then sent to a server, where a speech recognition algorithm is used to analyze the voice data and detect the words "busy" and "tired." A movement analysis algorithm is then used to detect unnatural hand movements.

[0182] Example prompts to input to a generative AI model:

[0183] Please explain a system that uses voice and motion data to analyze store staff stress in real time, adjust compensation, and integrate job information. Please particularly focus on the specific use of voice recognition and motion analysis algorithms.

[0184] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0185] Step 1:

[0186] The terminal collects voice and movement data from the employee via a wearable device. The collected voice data is obtained through a microphone, and movement data is acquired from an acceleration sensor. The input is the voice and movement data collected in real time, which is then transferred to a server. The terminal periodically sends the data to the server.

[0187] Step 2:

[0188] The server converts the received voice data into text using a speech recognition algorithm. This algorithm uses the Python "speech_recognition" library. The input is voice data and the output is text data. The analyzed text is checked to see if it contains keywords that indicate stress.

[0189] Step 3:

[0190] The server analyzes the received motion data using a motion analysis algorithm. This algorithm uses Python's "scipy.signal.find_peaks". The input is the motion data and the output is the number of motion events. If the motion events exceed a certain threshold, it is considered an indicator of stress.

[0191] Step 4:

[0192] The server quantifies the stress level based on the results of the analysis of the voice data and movement data. The stress level is calculated based on the number of detected keywords and the number of movement events. The input is the analysis results of the voice data and movement data, and the output is a stress score.

[0193] Step 5:

[0194] The server adjusts the employee's compensation based on the calculated stress score. The compensation adjustment algorithm adjusts the base salary according to the stress score and adds a bonus if necessary. The input is the stress score, and the output is the adjusted compensation amount.

[0195] Step 6:

[0196] The server connects the quantified stress score to the job information system, allowing job seekers to understand the working environment in advance. The input is the stress score, and the output is the information published in the job information system.

[0197] Step 7:

[0198] The server provides employees with a numerical stress score, allowing them to understand their own stress level in real time and consider ways to improve it. The input is the stress score, and the output is feedback information provided to employees.

[0199] 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.

[0200] This invention aims to visualize stress in nursing care settings and reflect it in compensation, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more accurate stress assessment. This system runs through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, providing information, and recognizing emotions.

[0201] Program processing

[0202] 1. Data Collection

[0203] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using a microphone, and movement data is measured using an accelerometer. This data is periodically transferred to a server.

[0204] 2. Collecting Emotional Data

[0205] The terminal uses sensors built into the wearable device to collect emotional data such as facial expressions and body movements, which are then transmitted to the server along with voice and movement data.

[0206] 3. Data Analysis

[0207] The server analyzes the received voice data using a speech recognition algorithm to identify the presence and frequency of abusive language, the behavioral analysis algorithm to detect violent behavior, and the emotion engine to analyze the user's emotional state.

[0208] 4. Quantification

[0209] The server calculates a stress score for each caregiver based on the analyzed voice, movement, and emotion data. This score is calculated based on certain criteria and can be aggregated weekly, monthly, or yearly.

[0210] 5. Remuneration adjustment

[0211] The server automatically adjusts the compensation of each care worker based on the numerical stress score. Specifically, it calculates the amount of bonuses and allowances according to the score and reflects this in the payroll system.

[0212] 6. Information provision

[0213] The server then links the calculated stress score to a job information system and makes it available for job seekers to view, allowing them to understand the working environment in advance and choose an appropriate workplace.

[0214] Specific examples

[0215] For example, let's say that caregiver Yamada is wearing a wearable device and encounters the following situation.

[0216] Monday

[0217] While working, Yamada was subjected to verbal abuse from users twice, such as being called "stupid" and "useless." The acceleration sensor also recorded one instance of Yamada being tapped on the shoulder. Furthermore, the facial expression recognition function detected that Yamada was under stress.

[0218] Tuesday

[0219] Yamada was verbally abused once, with the man telling her to "Stop it!", but no violence occurred. However, analysis of the tone of his voice revealed that he was in a state of serious stress.

[0220] The server receives this data and uses a voice recognition algorithm, behavioral analysis algorithm, and emotion engine to determine two cases of verbal abuse, one case of violence, and a high level of emotional stress, then quantifies the results and reflects them in the weekly score.

[0221] Based on the stress score obtained in this way, the server automatically adjusts Yamada's compensation and reflects it in the payroll system. At the same time, the quantified information is also linked to the job information system, allowing other job seekers to view and understand it in real time.

[0222] Through the above process, the present invention aims to reduce turnover by more accurately visualizing the stress of care workers and appropriately reflecting it in their compensation. The introduction of an emotion engine will further improve the working environment for care workers and support the recruitment and development of human resources throughout the care industry.

[0223] The processing flow will be explained below.

[0224] Step 1:

[0225] The user, a caregiver, puts on the wearable device and turns it on, and the device is automatically paired with the server.

[0226] Step 2:

[0227] The device collects real-time voice and movement data through the wearable device's microphone and accelerometer, as well as emotional data using facial expression and other emotion recognition sensors.

[0228] Step 3:

[0229] The device stores the collected voice data, movement data, and emotion data in a buffer at regular intervals (for example, every minute). Each data is assigned a timestamp.

[0230] Step 4:

[0231] The terminal transmits all the data in the buffer to the server at once. If a transmission error occurs, the terminal performs a retransmission process.

[0232] Step 5:

[0233] The server pre-processes the received audio data, including noise filtering and voice normalization, and also pre-processes the collected emotion data to remove outliers.

[0234] Step 6:

[0235] The server runs a speech recognition algorithm to convert the preprocessed audio data into a string of text, from which natural language processing (NLP) is used to extract abusive keywords.

[0236] Step 7:

[0237] The server pre-processes the motion data, filtering the accelerometer data to remove inaccurate data.

[0238] Step 8:

[0239] The server runs a behavioral analysis algorithm to detect patterns of violence based on sudden movements and impacts from the accelerometer data.

[0240] Step 9:

[0241] The server uses an emotion engine to analyze the user's emotional state from the pre-processed emotion data, including analyzing voice tone, speech rate, volume, facial expressions, and body movements.

[0242] Step 10:

[0243] The server integrates information extracted from the voice data, movement data, and emotion data to calculate a stress score for each caregiver, which is quantified based on a set of criteria.

[0244] Step 11:

[0245] The server aggregates the calculated stress scores on a weekly, monthly, and yearly basis, and automatically adjusts the compensation of each care worker based on this.

[0246] Step 12:

[0247] The server connects the results of the compensation adjustment to the payroll system and reflects them in the salaries of each care worker.

[0248] Step 13:

[0249] The server links the quantified stress score to a job information system and makes it publicly available for job seekers to view.

[0250] Step 14:

[0251] Users can browse job information pages and check the stress scores and working environments of each workplace to choose a workplace that suits them.

[0252] As a concrete example, let's say that care worker Yamada wears a wearable device and receives two verbal abuses from a caregiver on Monday ("idiot" and "useless"), and receives one slap on the shoulder. If facial expression recognition detects that Yamada is stressed, the server analyzes this data and reflects a quantified stress score in the payroll system and job information system. This process allows Yamada to receive fair compensation, and job seekers can obtain accurate information about the work environment.

[0253] Example 2

[0254] 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."

[0255] There is a need for a method to visualize and accurately evaluate the stress caused by verbal abuse and violence experienced by caregivers in the nursing care field. Furthermore, there is a need to improve the working environment and reduce staff turnover by appropriately adjusting the compensation of caregivers based on this stress assessment. Furthermore, there is a need for a method to provide information that allows job seekers to understand the working environment in advance.

[0256] 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.

[0257] In this invention, the server includes: means for collecting voice data and movement data received by caregivers wearing wearable devices; means for analyzing the collected voice data to determine whether verbal abuse has occurred; means for analyzing the collected movement data to determine whether violence has occurred; means for collecting emotional data such as facial expressions and body movements to determine the emotional state of the caregiver; means for analyzing the collected emotional data to determine the emotional state of the caregiver; means for quantifying the voice data, movement data, and emotional data based on the analysis results and calculating the caregiver's stress score; means for adjusting the caregiver's compensation based on the quantified information; and means for linking the quantified information to a job information system. This allows for accurate visualization of caregiver stress and adjustment of compensation based on the assessment, thereby improving the working environment and reducing staff turnover. It also allows job seekers to understand the working environment in advance.

[0258] A "wearable device" is a portable device designed to be worn by caregivers and equipped with sensors, microphones, cameras, etc. to collect voice, movement, and emotional data.

[0259] "Audio data" refers to electronically recorded audio information such as conversations that occur during the work of caregivers and environmental sounds.

[0260] "Motion data" refers to electronically recorded data relating to the physical movements and behavior of caregivers, and is collected using an acceleration sensor or the like.

[0261] "Emotional data" refers to electronically recorded data that indicates the emotional state of a caregiver, as inferred from facial expressions, physical movements, heart rate, and electrodermal activity.

[0262] A "voice recognition algorithm" is a technology that analyzes collected voice data, converts it into text data, and detects specific keywords and abusive language.

[0263] "Behavioral analysis algorithms" are technologies that analyze collected behavioral data to detect violent acts and abnormal behavior.

[0264] The "emotion engine" is a technology that analyzes collected emotional data and determines the emotional state of caregivers.

[0265] The "stress score" is a numerical representation of the level of stress of a caregiver calculated based on the results of analyzing voice data, movement data, and emotion data.

[0266] "Remuneration adjustment means" refers to technology and systems for automatically calculating and changing the remuneration of care workers based on stress scores.

[0267] A "job information system" is a system that provides information about working conditions that is made publicly available for job seekers to view.

[0268] This invention aims to visualize stress in nursing care settings and reflect it in compensation, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more accurate stress assessment. This system runs through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, providing information, and recognizing emotions.

[0269] Hardware and Software Configuration

[0270] The terminal uses a wearable device (e.g., a smartwatch or fitness tracker) to collect voice, movement, and emotion data of caregivers in real time. Voice data is collected using the device's built-in microphone, movement data is measured using an acceleration sensor, and emotion data is collected using facial recognition sensors and heart rate sensors.

[0271] The server receives the collected data and performs the following analysis:

[0272] 1. Analysis of audio data

[0273] The server converts the audio data into text using a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API), then detects specific keywords and abusive language.

[0274] 2. Analysis of behavioral data

[0275] The server analyzes the motion data using a behavioral analysis algorithm to detect violent acts and abnormal behavior.

[0276] 3. Emotion Data Analysis

[0277] The server analyzes the emotion data using an emotion engine (e.g., Google Cloud Natural Language API) and determines the emotional state of the caregiver.

[0278] Specific examples

[0279] Let's say that caregiver Yamada wears a wearable device and encounters the following situation.

[0280] Monday

[0281] The device collects data on two occasions when Yamada was subjected to abusive language such as "idiot" and "useless" from users while working.

[0282] The device uses an accelerometer to record the tapping of Yamada's shoulder once.

[0283] The device's facial expression recognition function detects when Yamada is expressing stress.

[0284] The server analyzes this data and determines that there are two cases of verbal abuse, one case of violence, and a high level of stress, and quantifies this data.

[0285] Tuesday

[0286] The device collects data that it received one verbal abuse such as "Stop it!", but records that no violence occurred.

[0287] The device analyzes the voice tone and detects that Yamada is in a state of serious stress.

[0288] From this data, the server determines that one incident of abusive language is a high level of stress and converts it into a numerical value.

[0289] Examples of prompt statements

[0290] An example of a prompt to be input to the generative AI model is as follows:

[0291] Please explain the system that calculates the stress score of caregivers based on the voice data, movement data, and emotion data collected from the caregivers' wearable devices, and automatically adjusts compensation based on that score. The voice data is used to check the presence and frequency of abusive language, and the movement data is used to identify violent acts. The emotion data is used to analyze the stress level using an emotion engine.

[0292] This system will accurately visualize the stress levels of care workers and adjust compensation based on the results, which will improve the working environment and reduce staff turnover. It will also enable job seekers to understand the working environment in advance.

[0293] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0294] Step 1: Data collection

[0295] The terminal collects voice data, movement data, and emotion data of the caregiver wearing the wearable device.

[0296] Input: Voice, movement, and facial expression generated by caregivers

[0297] Specific behavior:

[0298] The device uses a built-in microphone to collect audio data, such as the content of conversations and background sounds.

[0299] The device uses an integrated accelerometer to collect motion data, such as the number of steps taken by the caregiver, posture, and sudden movements.

[0300] The device collects emotional data using built-in facial recognition and heart rate sensors, for example, recording the caregiver's facial expressions and heart rate.

[0301] Output: Collected voice data, movement data, and emotion data

[0302] Step 2: Data Transfer

[0303] The terminal periodically transfers the collected data to the server.

[0304] Input: Collected voice data, movement data, emotion data

[0305] Specific behavior:

[0306] The terminal aggregates the collected data at regular intervals, encrypts it, and sends it to the server.

[0307] Output: Voice data, movement data, and emotion data transferred to the server

[0308] Step 3: Analyzing the audio data

[0309] The server analyzes the received voice data and determines whether or not abusive language is used.

[0310] Input: Audio data sent from the device

[0311] Specific behavior:

[0312] The server converts the voice data into text using a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API).

[0313] The server detects specific keywords and abusive language in the text data.

[0314] Output: Whether or not abusive language was detected and its frequency

[0315] Step 4: Analyzing the behavioral data

[0316] The server analyzes the received motion data and determines whether or not violence has occurred.

[0317] Input: Operation data sent from the device

[0318] Specific behavior:

[0319] The server uses a behavioral analysis algorithm to detect violent acts and abnormal behavior from the motion data.

[0320] For example, violent acts can be identified from abnormal acceleration values ​​or specific movement patterns.

[0321] Output: Whether or not violent acts were detected and their frequency

[0322] Step 5: Analyze the sentiment data

[0323] The server analyzes the received emotion data and determines the emotional state of the caregiver.

[0324] Input: Emotion data sent from the device

[0325] Specific behavior:

[0326] The server uses an emotion engine (e.g., Google Cloud Natural Language API) to classify the emotional state based on data from facial expression recognition sensors and heart rate sensors.

[0327] For example, it can recognize emotions such as smiles, anxiety, and anger from facial expressions, and estimate the level of stress from fluctuations in heart rate.

[0328] Output: Determined emotional state

[0329] Step 6: Calculate your stress score

[0330] The server calculates the stress score of the caregiver based on the analysis results.

[0331] Input: Analyzed voice data, movement data, emotion data

[0332] Specific behavior:

[0333] The server combines the information obtained from each data point and calculates a stress score, taking into account, for example, the number of abusive verbal and violent acts and changes in emotional state.

[0334] Output: Calculated stress score

[0335] Step 7: Compensation Adjustment

[0336] The server adjusts the caregiver's compensation based on the calculated stress score.

[0337] Input: Calculated stress score

[0338] Specific behavior:

[0339] The server calculates the amount of bonus or allowance based on the stress score.

[0340] The calculation results are linked to the payroll system and reflected in monthly pay slips.

[0341] Output: Adjusted reward information

[0342] Step 8: Provide information

[0343] The server links the stress score to a job information system and makes it publicly available for job seekers to view.

[0344] Input: Calculated stress score

[0345] Specific behavior:

[0346] The server uploads the stress score to the job information system and makes it public.

[0347] This allows job seekers to understand the working environment in advance.

[0348] Output: Published stress scores and work environment information

[0349] (Application example 2)

[0350] 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."

[0351] Conventional systems lacked a method for accurately assessing the stress levels of workers in nursing care and factory settings and directly reflecting the results in their compensation. In particular, stress assessments that took emotional data into account were not performed, creating challenges in improving the working environment and increasing worker motivation. Furthermore, there were also insufficient methods for visualizing workers' stress levels in detail and providing information that would allow job seekers to know the working environment in advance.

[0352] 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.

[0353] In this invention, the server includes means for collecting voice data and motion data received by workers wearing wearable devices, means for analyzing the collected voice data and determining whether abusive language has been used, means for analyzing the collected motion data and determining whether violence has been used, means for analyzing the emotion data using an emotion engine and identifying the level of stress, means for quantifying the level of abusive language, violence, and emotion data based on the analysis results, means for adjusting worker compensation based on the quantified information, and means for linking the quantified information to a job information system. This makes it possible to accurately assess worker stress and adjust compensation based on the assessment, as well as enable job seekers to understand the working environment in advance.

[0354] A "wearable device" is a device that can be worn by workers and collects voice data, movement data, emotional data, etc.

[0355] "Audio data" is digital data that records the words spoken by workers and the sounds of their surroundings.

[0356] "Movement data" is digital data collected by sensors on specific physical movements of workers.

[0357] "Abusive language" refers to words that have an insulting or negative connotation towards others.

[0358] "Violence" refers to the act of physically aggressing or inflicting harm on another person.

[0359] An "emotion engine" is software or algorithms that analyze the emotional state of workers based on collected data.

[0360] "Stress level" is a numerical representation of a worker's mental and emotional burden.

[0361] "Quantification" is the process of converting qualitative data into quantitative data that can be treated as such.

[0362] "Remuneration" refers to the monetary compensation received by workers, such as salary, bonuses, and allowances.

[0363] A "job information system" is an online platform where companies and facilities provide job information to job seekers.

[0364] The system for realizing this invention is composed of a wearable device, a server, and a job information system. The processing content of each step, the hardware and software used, and specific examples are explained in detail below.

[0365] Hardware and Software

[0366] Hardware

[0367] Wearable devices: Worn by workers, these devices include microphones to collect voice data, accelerometers to collect movement data, and cameras and sensors to recognize facial expressions.

[0368] Server: Used to analyze collected data and adjust rewards and share information.

[0369] Job Information System: An online platform that provides job seekers with a viewable stress score.

[0370] software

[0371] Speech recognition algorithm: Analyzes collected voice data to determine whether abusive language is present.

[0372] Motion analysis algorithm: Analyzes collected motion data to determine whether violence is present or not.

[0373] Emotion engine: Software for analyzing emotional data and identifying stress levels.

[0374] Compensation Adjustment System: Automatically adjusts workers' compensation based on quantified information.

[0375] Information linkage system: Software for linking quantified information with a job information system.

[0376] Processing steps and functions

[0377] The server processes data, adjusts rewards, and shares information based on the following steps:

[0378] 1. Data collection: Workers wear wearable devices to collect voice, movement, and emotion data, such as microphones, accelerometers, and facial recognition sensors.

[0379] 2. Data Transfer: Collected data is transferred to the server in real time or periodically.

[0380] 3. Data analysis: The server uses a voice recognition algorithm to analyze the voice data and determine whether or not there is abusive language. It also uses a motion analysis algorithm to identify whether or not there is violence. An emotion engine analyzes the level of stress from the emotional data.

[0381] 4. Quantification: Calculate the worker's stress score based on the analyzed voice, movement, and emotional data. For example, a comprehensive score is generated that combines the frequency of verbal and physical abuse with emotional data.

[0382] 5. Compensation Adjustment: Based on the score, the compensation adjustment system will automatically adjust the worker's compensation, which is calculated according to certain criteria.

[0383] 6. Information sharing: The quantified information is transferred to the employment information system via the information sharing system, where it can be viewed by job seekers.

[0384] Specific examples

[0385] For example, suppose a factory worker wears a wearable device while working and collects the following data:

[0386] Voice data: phrases like "I'm not feeling too great today" and stressed voice tones are collected.

[0387] Movement data: Occasionally, certain movement patterns are recorded that indicate strain or fatigue.

[0388] Emotional data: Cameras and sensors detect workers' haggard expressions.

[0389] These data are input into the generative AI model as example prompt sentences as follows:

[0390] Prompt Sentence Examples

[0391] Imagine that Sato is wearing a wearable device while working, and generate the following data.

[0392] Audio data: "I'm not feeling too good today," "I'll try a little harder."

[0393] Movement data: movement patterns that indicate strain or fatigue

[0394] Emotion data: Haggard expression

[0395] These processes enable accurate assessment of worker stress, appropriate compensation adjustments, and visualization of the working environment.

[0396] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0397] Step 1:

[0398] The terminal collects voice, movement, and emotion data from the worker wearing the wearable device. Specifically, it uses a microphone to capture voice data, an accelerometer to measure movement data, and a facial recognition sensor to collect emotion data. This data is then transferred to a server in real time or periodically.

[0399] Step 2:

[0400] The server analyzes the received voice data. Using a voice recognition algorithm, it analyzes specific keywords and tone of voice in the data to determine whether it contains abusive language. The input is the voice data, and the output is a flag indicating whether abusive language is present. For example, if the word "idiot" is included, it is detected as abusive language.

[0401] Step 3:

[0402] The server analyzes the received motion data. Using a motion analysis algorithm, it analyzes specific motion patterns and determines whether a violent act has occurred. The input is the motion data, and the output is a flag indicating whether a violent act has occurred. For example, if a rapid change in acceleration is recorded, it is detected as a violent act.

[0403] Step 4:

[0404] The server uses an emotion engine to analyze the emotion data and calculate the stress level. The input is emotion data, and the output is a numerical value (stress score) that indicates the emotional state. Specifically, if haggard expressions are frequently detected, the stress score will be high.

[0405] Step 5:

[0406] The server integrates the results of the analysis of the voice data, movement data, and emotion data to calculate an overall stress score. Each piece of data is weighted to calculate the overall score. For example, the overall score is created by averaging the frequency of abusive language, the frequency of violent behavior, and emotion data indicating high stress. The inputs are the results of each analysis, and the output is the overall stress score.

[0407] Step 6:

[0408] The server automatically adjusts the worker's compensation based on the calculated overall stress score. It determines whether to increase or decrease compensation depending on the score and transfers the result to the payroll system. Specifically, if the overall score is high, bonuses and allowances are increased. The input is the overall stress score, and the output is the amount of compensation.

[0409] Step 7:

[0410] The server links the quantified stress score to the job information system. The information is made available to job seekers in a viewable format, allowing them to understand the working environment in advance. Specifically, the server displays the stress score for each workplace, helping job seekers select the appropriate workplace. The input is the overall stress score, and the output is the information displayed on the job information system.

[0411] 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.

[0412] 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.

[0413] 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.

[0414] [Second embodiment]

[0415] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0416] 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.

[0417] 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).

[0418] 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.

[0419] 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.

[0420] 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).

[0421] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0422] 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.

[0423] 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.

[0424] 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.

[0425] 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.

[0426] 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."

[0427] This invention is a system for visualizing stress in nursing care settings and reflecting it in compensation. This system is implemented through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, and providing information.

[0428] Program processing

[0429] 1. Data Collection

[0430] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using a microphone, and movement data is measured using an accelerometer. This data is periodically transferred to a server.

[0431] 2. Data Analysis

[0432] The server analyzes the received voice data using a voice recognition algorithm to identify the presence and frequency of abusive language, and uses a behavior analysis algorithm to detect violent behavior from the movement data. This increases the detection rate of abusive language and violence, and allows for more accurate data to be obtained.

[0433] 3. Quantification

[0434] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. This score is calculated based on a set of criteria and can be aggregated weekly, monthly, or yearly.

[0435] 4. Remuneration adjustment

[0436] The server automatically adjusts the compensation of each care worker based on the numerical stress score. Specifically, it calculates the amount of bonuses and allowances according to the score and reflects this in the payroll system.

[0437] 5. Information provision

[0438] The server then links the quantified stress scores to a job information system and makes them available to job seekers, allowing them to understand the working environment in advance and choose an appropriate workplace.

[0439] Specific examples

[0440] For example, let's say that caregiver Sato is wearing a wearable device and encounters the following situation.

[0441] Monday

[0442] While working, Sato was subjected to verbal abuse from users, such as "idiot" and "useless," twice. The accelerometer also recorded him being hit in the chest once.

[0443] Tuesday

[0444] Sato was subjected to one verbal abuse, "Stop it!", but no violence occurred.

[0445] The server receives this data and uses voice recognition and behavioral analysis algorithms to determine two cases of verbal abuse and one case of violence, then quantifies the results and reflects them in the weekly score.

[0446] Based on the stress score obtained in this way, the server automatically adjusts Sato's compensation and reflects it in the payroll system. At the same time, this quantified information is also linked to the job information system, allowing other job seekers to view and understand it in real time.

[0447] Through the above process, the present invention aims to reduce turnover by visualizing the stress of care workers and appropriately reflecting this in their compensation. This system will improve the working environment for care workers and support the recruitment and development of human resources throughout the care industry.

[0448] The processing flow will be explained below.

[0449] Step 1:

[0450] The user, a caregiver, puts on the wearable device and turns it on, and the device is automatically paired with the server.

[0451] Step 2:

[0452] The device collects real-time audio data through the wearable device's microphone, while also collecting body movement data using an accelerometer.

[0453] Step 3:

[0454] The device stores the collected voice data and motion data in a buffer at regular intervals (for example, every minute). The stored data is given a timestamp.

[0455] Step 4:

[0456] The terminal transmits all the data in the buffer to the server at once. If a transmission error occurs, the terminal performs a retransmission process.

[0457] Step 5:

[0458] The server performs pre-processing of the received audio data, which includes noise filtering and voice normalization.

[0459] Step 6:

[0460] The server runs a speech recognition algorithm to convert the preprocessed audio data into text, from which it uses natural language processing (NLP) to extract abusive keywords.

[0461] Step 7:

[0462] The server pre-processes the motion data, filtering the accelerometer data to remove inaccurate data.

[0463] Step 8:

[0464] The server runs a behavioral analysis algorithm to detect patterns of violence in the filtered motion data, based on sudden movements and impacts.

[0465] Step 9:

[0466] The server calculates a stress score for each caregiver based on the extracted abusive keywords and the detected patterns of violence. The score is calculated based on certain criteria.

[0467] Step 10:

[0468] The server aggregates weekly, monthly, and yearly stress scores and adjusts compensation accordingly, which is then linked to the payroll system.

[0469] Step 11:

[0470] The server connects the calculated stress score to the job information system and makes it publicly available for job seekers to view.

[0471] Step 12:

[0472] Users can browse job information pages and check the stress scores of each workplace to choose a workplace that suits them.

[0473] Example 1

[0474] 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."

[0475] In the field of nursing care, caregivers often experience high levels of stress on a daily basis, and are frequently subjected to verbal and physical abuse. This leads to a decline in staff motivation and a high turnover rate. However, there is no way to objectively grasp the actual level of stress felt by caregivers, and appropriate compensation adjustments are not being made. To solve this issue, there is a need for a system that can quantify the stress of caregivers and automatically adjust compensation based on this information.

[0476] 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.

[0477] In this invention, the server includes means for collecting voice data and movement data received by caregivers wearing wearable devices, means for analyzing the collected voice data and determining whether abusive language has been used, means for analyzing the collected movement data and determining whether violence has been used, means for quantifying the degree of abusive language and violence based on the analysis results, means for automatically adjusting the caregiver's compensation based on the quantified information, means for linking the quantified information to a recruitment system and providing it, means for collecting voice data using an acoustic sensor and movement data using a motion sensor, and means for aggregating the quantified information on a weekly, monthly, and yearly basis and automatically adjusting compensation based on the quantified information. This makes it possible to objectively quantify the stress of caregivers and adjust compensation appropriately.

[0478] A "wearable device" is an electronic device that is worn on the body and is used to collect voice data and movement data.

[0479] "Voice data" refers to the audio information received by caregivers, and is digital data collected using microphones and acoustic sensors.

[0480] "Motion data" refers to information that measures the physical movements and acceleration of care workers, and is digital data collected using acceleration sensors and motion sensors.

[0481] "Analysis" is the process of using collected data to extract specific information and make decisions based on that information.

[0482] "Abusive language" refers to insulting words or expressions directed at nursing care staff.

[0483] "Violence" refers to physical aggression or force used against care staff.

[0484] "Quantification" is the process of expressing analyzed data as quantitative numbers.

[0485] "Automatic compensation adjustment" is a system that automatically calculates and adjusts the salaries and bonuses of care workers based on quantified information.

[0486] The "recruitment system" refers to an online platform where job seekers can view information about nursing care workplaces and apply.

[0487] An "acoustic sensor" is a device that detects sound information and collects it as digital data.

[0488] A "motion sensor" is a device that detects body movements and collects them as digital data.

[0489] This invention is a system for visualizing stress in nursing care settings and reflecting it in compensation. This system is implemented through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, and providing information.

[0490] 1. Data Collection

[0491] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using an acoustic sensor (microphone), and movement data is measured using a motion sensor (accelerometer). This data is periodically transferred to a server.

[0492] 2. Data Transfer

[0493] The collected voice and movement data is transferred to a server in real time via Wi-Fi or Bluetooth, for example, every minute, split into packets and sent with error checking to prevent data loss.

[0494] 3. Data Analysis

[0495] The server converts the received audio data into text using the Google Cloud Speech-to-Text API and uses natural language processing (NLP) algorithms to identify the presence and frequency of abusive language. The movement data is analyzed using a custom behavioral analysis model using TensorFlow to detect physical violence. For example, certain keywords or high accelerations can be detected to identify abusive language or violence.

[0496] 4. Quantification

[0497] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. The quantification criteria are 10 points for each verbal abuse incident and 50 points for each physical abuse incident, and the data can be compiled weekly, monthly, or yearly.

[0498] 5. Remuneration adjustment

[0499] The server automatically adjusts the compensation of each care worker based on the quantified stress score. Specifically, it calculates the amount of bonuses and allowances according to the stress score and reflects this in the payroll system (e.g., SAP SuccessFactors). For example, it can be set up so that an additional bonus is paid to staff whose monthly score exceeds 300 points.

[0500] 6. Information provision

[0501] The server connects the quantified stress score to recruitment systems (e.g., Recruit and Indeed) so that job seekers can understand the working environment in advance. For example, if Mr. Sato has a high stress score, that information will be displayed on the recruitment information page, allowing other job seekers to use it as a reference when choosing a workplace.

[0502] Specific examples

[0503] For example, let's say that caregiver Sato is wearing a wearable device and encounters the following situation.

[0504] Monday

[0505] While working, Sato was twice subjected to abusive language from users, such as being called an idiot and being useless, and an accelerometer recorded him being hit in the chest once.

[0506] Tuesday

[0507] Sato was subjected to one verbal abuse of "Stop it!" but no violence occurred.

[0508] The server receives this data and, using a custom behavioral analysis model using the Google Cloud Speech-to-Text API and TensorFlow, determines two instances of verbal abuse and one instance of violence, converts the results into a numerical value, and reflects this in a weekly score. Based on the stress score obtained in this way, the server automatically adjusts Mr. Sato's compensation and reflects it in the payroll system (e.g., SAP SuccessFactors). At the same time, this numerical information is also linked to job information systems (e.g., Recruit and Indeed), allowing other job seekers to view and understand it in real time.

[0509] Example prompts to input to the generative AI model

[0510] Prompt: Using the data below, explain the specific steps for a system that calculates stress scores for caregivers and automatically adjusts compensation based on the results.

[0511] Monday: Sato was subjected to verbal abuse such as "idiot" and "useless" twice during work hours, and an accelerometer recorded him being hit in the chest once.

[0512] Tuesday: One verbal abuse of "Stop it!", no violence occurred.

[0513] The system proceeds as follows:

[0514] 1. Data collection (collection of voice data and movement data, transfer to server)

[0515] 2. Data analysis (voice recognition and behavior analysis)

[0516] 3. Quantification (calculating stress scores)

[0517] 4. Compensation adjustment (automatic compensation adjustment based on numerical data)

[0518] 5. Information provision (linking to recruitment information systems)

[0519] This invention aims to reduce turnover by visualizing stress in the nursing care field and appropriately reflecting it in compensation. This system will improve the working environment for nursing care workers and support the recruitment and development of human resources throughout the nursing care industry.

[0520] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0521] Step 1: Data collection

[0522] The terminal starts up the wearable device worn by the caregiver and starts collecting voice and movement data. Voice data is collected using an acoustic sensor, and movement data is measured using a motion sensor. For example, voice data (such as "idiot" or "useless") is collected here, and the acceleration sensor detects the action of hitting the chest. The input of this step is the voice and movement of the caregiver, and the output is the collected raw data.

[0523] Step 2: Data Transfer

[0524] The device transfers the collected voice and movement data to the server in real time via Wi-Fi or Bluetooth. For example, the voice and movement data is compressed every minute and sent to the server in packet format, minimizing data loss. The input of this step is the collected raw data, and the output is the raw data stored on the server.

[0525] Step 3: Data analysis

[0526] The server converts the received voice data into text using the Google Cloud Speech-to-Text API and analyzes the presence and frequency of abusive language using natural language processing (NLP) algorithms. It also analyzes the behavioral data using a custom behavioral analysis model using TensorFlow to identify violent behavior. For example, it detects keywords (e.g., "idiot" or "stop") or high accelerations in the converted text. The input of this step is the received raw data, and the output is the analysis results.

[0527] Step 4: Quantify

[0528] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. For example, points are assigned, such as 10 points for each verbal abuse and 50 points for each physical abuse, and the totals are used to calculate the stress score. The input for this step is the analysis results, and the output is the calculated stress score.

[0529] Step 5: Compensation Adjustment

[0530] The server automatically adjusts the compensation of each caregiver based on the quantified stress score. Specifically, it calculates the amount of bonuses and allowances according to the stress score and connects this to the payroll system. For example, if the monthly score exceeds 300 points, a bonus is added. The input to this step is the calculated stress score, and the output is the adjusted compensation information.

[0531] Step 6: Provide information

[0532] The server links the quantified stress score to the job information system and provides the data so that job seekers can view it. For example, a job seeker can check the stress environment of Mr. Sato's workplace on the job information page. The input of this step is the adjusted compensation information, and the output is the stress score reflected in the job information system.

[0533] Through these steps, this system will improve the working environment in nursing care settings, appropriately reflect the stress of nursing care staff in their compensation, and enable job seekers to understand the working environment in advance.

[0534] (Application example 1)

[0535] 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."

[0536] Currently, in many brick-and-mortar stores, there are many situations where staff members who serve customers feel stressed. When this stress increases, employee motivation decreases and it may even lead to an increase in turnover. Furthermore, in order to improve customer satisfaction, it is important to create an environment where staff can work comfortably, but there is a lack of means to achieve this. Furthermore, there is a need to introduce an appropriate compensation system to make staff stress visible and improve the working environment.

[0537] 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.

[0538] In this invention, the server includes means for collecting voice data and movement data received by workers wearing wearable devices, means for analyzing the collected voice data and determining whether or not there are words that cause stress, means for analyzing the collected movement data and determining whether or not there are movement events that indicate stress, means for quantifying the stress level based on the analysis results, means for adjusting the worker's remuneration based on the quantified information, means for linking the quantified information to a recruitment information system, and means for providing the quantified information to the worker. This makes it possible to grasp staff stress in real time, adjust remuneration based on that information, and provide a comfortable working environment.

[0539] A "wearable device" is a device that can be worn on the body and collects voice data and movement data in real time.

[0540] "Employee" refers to a worker who provides customer service in a physical store.

[0541] "Audio data" refers to audio signals collected using a microphone and recorded as digital data.

[0542] "Motion data" is digital data indicating body movements obtained using an acceleration sensor or other sensor device.

[0543] "Stressor words" refer to specific words or phrases that are important when assessing an employee's stress level, and are often negative comments from customers or words that indicate overwork.

[0544] "Movement event" refers to a specific physical movement detected using sensors, and may include movements that are indicative of stress.

[0545] The "stress level" is the estimated level of stress felt by the worker, calculated from the collected voice data and movement data.

[0546] "Remuneration" refers to the monetary compensation, including salary and bonuses, that an employee receives for their work.

[0547] A "job information system" is an online system that handles information on the labor market and provides job seekers with data on companies and jobs.

[0548] "Means of providing" refers to methods or systems for communicating collected data and calculated results to employees, such as notification systems or portal sites.

[0549] MODE FOR CARRYING OUT THE INVENTION

[0550] To realize this invention, store employees who serve customers must wear a wearable device. The wearable device is used to collect voice and motion data in real time. The collected data is sent to a server and analyzed using voice recognition and motion analysis algorithms.

[0551] A description of what the program does

[0552] Hardware and software configuration:

[0553] Hardware:

[0554] Wearable device: Equipped with a microphone and accelerometer to collect voice and movement data.

[0555] Accelerometer: A sensor that detects body movement.

[0556] Microphone: A device for collecting sound.

[0557] software:

[0558] Server: A computer system that performs analysis and stores data.

[0559] Speech recognition algorithms: convert collected voice data into text and detect specific stressful words.

[0560] Motion analysis algorithm: Analyzes acceleration data to detect motion events that indicate stress. As a concrete example, peak detection is performed using the Python library "scipy."

[0561] Data processing and calculations:

[0562] The voice data is converted into text by a speech recognition algorithm on the server, which checks whether certain keywords (e.g., "busy," "tired," "terrible," etc.) are included to determine the stress factor.

[0563] Movement data is analyzed based on the output of the acceleration sensor to detect movement events that are indicators of stress (e.g., large or unnatural movements).

[0564] These analysis results are quantified and the employee's stress score is calculated.

[0565] Information provision and compensation adjustment:

[0566] The quantified information is stored on a server and compiled periodically, weekly, monthly, and annually, and employee compensation is automatically adjusted based on the results.

[0567] This information will also be linked to the job information system so that job seekers can view it.

[0568] The collected and analyzed stress information is also provided to employees, helping them understand their own stress levels and consider ways to improve them.

[0569] Examples:

[0570] For example, if a customer service staff member at a brick-and-mortar store hears a comment from a customer that they are "busy" or "tired" and begins to feel stressed during their shift, the voice data is collected through a microphone. An accelerometer is used to simultaneously detect movement events indicative of stress, such as hand vibrations. This data is then sent to a server, where a speech recognition algorithm is used to analyze the voice data and detect the words "busy" and "tired." A movement analysis algorithm is then used to detect unnatural hand movements.

[0571] Example prompts to input to a generative AI model:

[0572] Please explain a system that uses voice and motion data to analyze store staff stress in real time, adjust compensation, and integrate job information. Please particularly focus on the specific use of voice recognition and motion analysis algorithms.

[0573] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0574] Step 1:

[0575] The terminal collects voice and movement data from the employee via a wearable device. The collected voice data is obtained through a microphone, and movement data is acquired from an acceleration sensor. The input is the voice and movement data collected in real time, which is then transferred to a server. The terminal periodically sends the data to the server.

[0576] Step 2:

[0577] The server converts the received voice data into text using a speech recognition algorithm. This algorithm uses the Python "speech_recognition" library. The input is voice data and the output is text data. The analyzed text is checked to see if it contains keywords that indicate stress.

[0578] Step 3:

[0579] The server analyzes the received motion data using a motion analysis algorithm. This algorithm uses Python's "scipy.signal.find_peaks". The input is the motion data and the output is the number of motion events. If the motion events exceed a certain threshold, it is considered an indicator of stress.

[0580] Step 4:

[0581] The server quantifies the stress level based on the results of the analysis of the voice data and movement data. The stress level is calculated based on the number of detected keywords and the number of movement events. The input is the analysis results of the voice data and movement data, and the output is a stress score.

[0582] Step 5:

[0583] The server adjusts the employee's compensation based on the calculated stress score. The compensation adjustment algorithm adjusts the base salary according to the stress score and adds a bonus if necessary. The input is the stress score, and the output is the adjusted compensation amount.

[0584] Step 6:

[0585] The server connects the quantified stress score to the job information system, allowing job seekers to understand the working environment in advance. The input is the stress score, and the output is the information published in the job information system.

[0586] Step 7:

[0587] The server provides employees with a numerical stress score, allowing them to understand their own stress level in real time and consider ways to improve it. The input is the stress score, and the output is feedback information provided to employees.

[0588] 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.

[0589] This invention aims to visualize stress in nursing care settings and reflect it in compensation, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more accurate stress assessment. This system runs through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, providing information, and recognizing emotions.

[0590] Program processing

[0591] 1. Data Collection

[0592] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using a microphone, and movement data is measured using an accelerometer. This data is periodically transferred to a server.

[0593] 2. Collecting Emotional Data

[0594] The terminal uses sensors built into the wearable device to collect emotional data such as facial expressions and body movements, which are then transmitted to the server along with voice and movement data.

[0595] 3. Data Analysis

[0596] The server analyzes the received voice data using a speech recognition algorithm to identify the presence and frequency of abusive language, the behavioral analysis algorithm to detect violent behavior, and the emotion engine to analyze the user's emotional state.

[0597] 4. Quantification

[0598] The server calculates a stress score for each caregiver based on the analyzed voice, movement, and emotion data. This score is calculated based on certain criteria and can be aggregated weekly, monthly, or yearly.

[0599] 5. Remuneration adjustment

[0600] The server automatically adjusts the compensation of each care worker based on the numerical stress score. Specifically, it calculates the amount of bonuses and allowances according to the score and reflects this in the payroll system.

[0601] 6. Information provision

[0602] The server then links the calculated stress score to a job information system and makes it available for job seekers to view, allowing them to understand the working environment in advance and choose an appropriate workplace.

[0603] Specific examples

[0604] For example, let's say that caregiver Yamada is wearing a wearable device and encounters the following situation.

[0605] Monday

[0606] While working, Yamada was subjected to verbal abuse from users twice, such as being called "stupid" and "useless." The acceleration sensor also recorded one instance of Yamada being tapped on the shoulder. Furthermore, the facial expression recognition function detected that Yamada was under stress.

[0607] Tuesday

[0608] Yamada was verbally abused once, with the man telling her to "Stop it!", but no violence occurred. However, analysis of the tone of his voice revealed that he was in a state of serious stress.

[0609] The server receives this data and uses a voice recognition algorithm, behavioral analysis algorithm, and emotion engine to determine two cases of verbal abuse, one case of violence, and a high level of emotional stress, then quantifies the results and reflects them in the weekly score.

[0610] Based on the stress score obtained in this way, the server automatically adjusts Yamada's compensation and reflects it in the payroll system. At the same time, the quantified information is also linked to the job information system, allowing other job seekers to view and understand it in real time.

[0611] Through the above process, the present invention aims to reduce turnover by more accurately visualizing the stress of care workers and appropriately reflecting it in their compensation. The introduction of an emotion engine will further improve the working environment for care workers and support the recruitment and development of human resources throughout the care industry.

[0612] The processing flow will be explained below.

[0613] Step 1:

[0614] The user, a caregiver, puts on the wearable device and turns it on, and the device is automatically paired with the server.

[0615] Step 2:

[0616] The device collects real-time voice and movement data through the wearable device's microphone and accelerometer, as well as emotional data using facial expression and other emotion recognition sensors.

[0617] Step 3:

[0618] The device stores the collected voice data, movement data, and emotion data in a buffer at regular intervals (for example, every minute). Each data is assigned a timestamp.

[0619] Step 4:

[0620] The terminal transmits all the data in the buffer to the server at once. If a transmission error occurs, the terminal performs a retransmission process.

[0621] Step 5:

[0622] The server pre-processes the received audio data, including noise filtering and voice normalization, and also pre-processes the collected emotion data to remove outliers.

[0623] Step 6:

[0624] The server runs a speech recognition algorithm to convert the preprocessed audio data into a string of text, from which natural language processing (NLP) is used to extract abusive keywords.

[0625] Step 7:

[0626] The server pre-processes the motion data, filtering the accelerometer data to remove inaccurate data.

[0627] Step 8:

[0628] The server runs a behavioral analysis algorithm to detect patterns of violence based on sudden movements and impacts from the accelerometer data.

[0629] Step 9:

[0630] The server uses an emotion engine to analyze the user's emotional state from the pre-processed emotion data, including analyzing voice tone, speech rate, volume, facial expressions, and body movements.

[0631] Step 10:

[0632] The server integrates information extracted from the voice data, movement data, and emotion data to calculate a stress score for each caregiver, which is quantified based on a set of criteria.

[0633] Step 11:

[0634] The server aggregates the calculated stress scores on a weekly, monthly, and yearly basis, and automatically adjusts the compensation of each care worker based on this.

[0635] Step 12:

[0636] The server connects the results of the compensation adjustment to the payroll system and reflects them in the salaries of each care worker.

[0637] Step 13:

[0638] The server links the quantified stress score to a job information system and makes it publicly available for job seekers to view.

[0639] Step 14:

[0640] Users can browse job information pages and check the stress scores and working environments of each workplace to choose a workplace that suits them.

[0641] As a concrete example, let's say that care worker Yamada wears a wearable device and receives two verbal abuses from a caregiver on Monday ("idiot" and "useless"), and receives one slap on the shoulder. If facial expression recognition detects that Yamada is stressed, the server analyzes this data and reflects a quantified stress score in the payroll system and job information system. This process allows Yamada to receive fair compensation, and job seekers can obtain accurate information about the work environment.

[0642] Example 2

[0643] 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."

[0644] There is a need for a method to visualize and accurately evaluate the stress caused by verbal abuse and violence experienced by caregivers in the nursing care field. Furthermore, there is a need to improve the working environment and reduce staff turnover by appropriately adjusting the compensation of caregivers based on this stress assessment. Furthermore, there is a need for a method to provide information that allows job seekers to understand the working environment in advance.

[0645] 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.

[0646] In this invention, the server includes: means for collecting voice data and movement data received by caregivers wearing wearable devices; means for analyzing the collected voice data to determine whether verbal abuse has occurred; means for analyzing the collected movement data to determine whether violence has occurred; means for collecting emotional data such as facial expressions and body movements to determine the emotional state of the caregiver; means for analyzing the collected emotional data to determine the emotional state of the caregiver; means for quantifying the voice data, movement data, and emotional data based on the analysis results and calculating the caregiver's stress score; means for adjusting the caregiver's compensation based on the quantified information; and means for linking the quantified information to a job information system. This allows for accurate visualization of caregiver stress and adjustment of compensation based on the assessment, thereby improving the working environment and reducing staff turnover. It also allows job seekers to understand the working environment in advance.

[0647] A "wearable device" is a portable device designed to be worn by caregivers and equipped with sensors, microphones, cameras, etc. to collect voice, movement, and emotional data.

[0648] "Audio data" refers to electronically recorded audio information such as conversations that occur during the work of caregivers and environmental sounds.

[0649] "Motion data" refers to electronically recorded data relating to the physical movements and behavior of caregivers, and is collected using an acceleration sensor or the like.

[0650] "Emotional data" refers to electronically recorded data that indicates the emotional state of a caregiver, as inferred from facial expressions, physical movements, heart rate, and electrodermal activity.

[0651] A "voice recognition algorithm" is a technology that analyzes collected voice data, converts it into text data, and detects specific keywords and abusive language.

[0652] "Behavioral analysis algorithms" are technologies that analyze collected behavioral data to detect violent acts and abnormal behavior.

[0653] The "emotion engine" is a technology that analyzes collected emotional data and determines the emotional state of caregivers.

[0654] The "stress score" is a numerical representation of the level of stress of a caregiver calculated based on the results of analyzing voice data, movement data, and emotion data.

[0655] "Remuneration adjustment means" refers to technology and systems for automatically calculating and changing the remuneration of care workers based on stress scores.

[0656] A "job information system" is a system that provides information about working conditions that is made publicly available for job seekers to view.

[0657] This invention aims to visualize stress in nursing care settings and reflect it in compensation, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more accurate stress assessment. This system runs through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, providing information, and recognizing emotions.

[0658] Hardware and Software Configuration

[0659] The terminal uses a wearable device (e.g., a smartwatch or fitness tracker) to collect voice, movement, and emotion data of caregivers in real time. Voice data is collected using the device's built-in microphone, movement data is measured using an acceleration sensor, and emotion data is collected using facial recognition sensors and heart rate sensors.

[0660] The server receives the collected data and performs the following analysis:

[0661] 1. Analysis of audio data

[0662] The server converts the audio data into text using a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API), then detects specific keywords and abusive language.

[0663] 2. Analysis of behavioral data

[0664] The server analyzes the motion data using a behavioral analysis algorithm to detect violent acts and abnormal behavior.

[0665] 3. Emotion Data Analysis

[0666] The server analyzes the emotion data using an emotion engine (e.g., Google Cloud Natural Language API) and determines the emotional state of the caregiver.

[0667] Specific examples

[0668] Let's say that caregiver Yamada wears a wearable device and encounters the following situation.

[0669] Monday

[0670] The device collects data on two occasions when Yamada was subjected to abusive language such as "idiot" and "useless" from users while working.

[0671] The device uses an accelerometer to record the tapping of Yamada's shoulder once.

[0672] The device's facial expression recognition function detects when Yamada is expressing stress.

[0673] The server analyzes this data and determines that there are two cases of verbal abuse, one case of violence, and a high level of stress, and quantifies this data.

[0674] Tuesday

[0675] The device collects data that it received one verbal abuse such as "Stop it!", but records that no violence occurred.

[0676] The device analyzes the voice tone and detects that Yamada is in a state of serious stress.

[0677] From this data, the server determines that one incident of abusive language is a high level of stress and converts it into a numerical value.

[0678] Examples of prompt statements

[0679] An example of a prompt to be input to the generative AI model is as follows:

[0680] Please explain the system that calculates the stress score of caregivers based on the voice data, movement data, and emotion data collected from the caregivers' wearable devices, and automatically adjusts compensation based on that score. The voice data is used to check the presence and frequency of abusive language, and the movement data is used to identify violent acts. The emotion data is used to analyze the stress level using an emotion engine.

[0681] This system will accurately visualize the stress levels of care workers and adjust compensation based on the results, which will improve the working environment and reduce staff turnover. It will also enable job seekers to understand the working environment in advance.

[0682] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0683] Step 1: Data collection

[0684] The terminal collects voice data, movement data, and emotion data of the caregiver wearing the wearable device.

[0685] Input: Voice, movement, and facial expression generated by caregivers

[0686] Specific behavior:

[0687] The device uses a built-in microphone to collect audio data, such as the content of conversations and background sounds.

[0688] The device uses an integrated accelerometer to collect motion data, such as the number of steps taken by the caregiver, posture, and sudden movements.

[0689] The device collects emotional data using built-in facial recognition and heart rate sensors, for example, recording the caregiver's facial expressions and heart rate.

[0690] Output: Collected voice data, movement data, and emotion data

[0691] Step 2: Data Transfer

[0692] The terminal periodically transfers the collected data to the server.

[0693] Input: Collected voice data, movement data, emotion data

[0694] Specific behavior:

[0695] The terminal aggregates the collected data at regular intervals, encrypts it, and sends it to the server.

[0696] Output: Voice data, movement data, and emotion data transferred to the server

[0697] Step 3: Analyzing the audio data

[0698] The server analyzes the received voice data and determines whether or not abusive language is used.

[0699] Input: Audio data sent from the device

[0700] Specific behavior:

[0701] The server converts the voice data into text using a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API).

[0702] The server detects specific keywords and abusive language in the text data.

[0703] Output: Whether or not abusive language was detected and its frequency

[0704] Step 4: Analyzing the behavioral data

[0705] The server analyzes the received motion data and determines whether or not violence has occurred.

[0706] Input: Operation data sent from the device

[0707] Specific behavior:

[0708] The server uses a behavioral analysis algorithm to detect violent acts and abnormal behavior from the motion data.

[0709] For example, violent acts can be identified from abnormal acceleration values ​​or specific movement patterns.

[0710] Output: Whether or not violent acts were detected and their frequency

[0711] Step 5: Analyze the sentiment data

[0712] The server analyzes the received emotion data and determines the emotional state of the caregiver.

[0713] Input: Emotion data sent from the device

[0714] Specific behavior:

[0715] The server uses an emotion engine (e.g., Google Cloud Natural Language API) to classify the emotional state based on data from facial expression recognition sensors and heart rate sensors.

[0716] For example, it can recognize emotions such as smiles, anxiety, and anger from facial expressions, and estimate the level of stress from fluctuations in heart rate.

[0717] Output: Determined emotional state

[0718] Step 6: Calculate your stress score

[0719] The server calculates the stress score of the caregiver based on the analysis results.

[0720] Input: Analyzed voice data, movement data, emotion data

[0721] Specific behavior:

[0722] The server combines the information obtained from each data point and calculates a stress score, taking into account, for example, the number of abusive verbal and violent acts and changes in emotional state.

[0723] Output: Calculated stress score

[0724] Step 7: Compensation Adjustment

[0725] The server adjusts the caregiver's compensation based on the calculated stress score.

[0726] Input: Calculated stress score

[0727] Specific behavior:

[0728] The server calculates the amount of bonus or allowance based on the stress score.

[0729] The calculation results are linked to the payroll system and reflected in monthly pay slips.

[0730] Output: Adjusted reward information

[0731] Step 8: Provide information

[0732] The server links the stress score to a job information system and makes it publicly available for job seekers to view.

[0733] Input: Calculated stress score

[0734] Specific behavior:

[0735] The server uploads the stress score to the job information system and makes it public.

[0736] This allows job seekers to understand the working environment in advance.

[0737] Output: Published stress scores and work environment information

[0738] (Application example 2)

[0739] 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."

[0740] Conventional systems lacked a method for accurately assessing the stress levels of workers in nursing care and factory settings and directly reflecting the results in their compensation. In particular, stress assessments that took emotional data into account were not performed, creating challenges in improving the working environment and increasing worker motivation. Furthermore, there were also insufficient methods for visualizing workers' stress levels in detail and providing information that would allow job seekers to know the working environment in advance.

[0741] 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.

[0742] In this invention, the server includes means for collecting voice data and motion data received by workers wearing wearable devices, means for analyzing the collected voice data and determining whether abusive language has been used, means for analyzing the collected motion data and determining whether violence has been used, means for analyzing the emotion data using an emotion engine and identifying the level of stress, means for quantifying the level of abusive language, violence, and emotion data based on the analysis results, means for adjusting worker compensation based on the quantified information, and means for linking the quantified information to a job information system. This makes it possible to accurately assess worker stress and adjust compensation based on the assessment, as well as enable job seekers to understand the working environment in advance.

[0743] A "wearable device" is a device that can be worn by workers and collects voice data, movement data, emotional data, etc.

[0744] "Audio data" is digital data that records the words spoken by workers and the sounds of their surroundings.

[0745] "Movement data" is digital data collected by sensors on specific physical movements of workers.

[0746] "Abusive language" refers to words that have an insulting or negative connotation towards others.

[0747] "Violence" refers to the act of physically aggressing or inflicting harm on another person.

[0748] An "emotion engine" is software or algorithms that analyze the emotional state of workers based on collected data.

[0749] "Stress level" is a numerical representation of a worker's mental and emotional burden.

[0750] "Quantification" is the process of converting qualitative data into quantitative data that can be treated as such.

[0751] "Remuneration" refers to the monetary compensation received by workers, such as salary, bonuses, and allowances.

[0752] A "job information system" is an online platform where companies and facilities provide job information to job seekers.

[0753] The system for realizing this invention is composed of a wearable device, a server, and a job information system. The processing content of each step, the hardware and software used, and specific examples are explained in detail below.

[0754] Hardware and Software

[0755] Hardware

[0756] Wearable devices: Worn by workers, these devices include microphones to collect voice data, accelerometers to collect movement data, and cameras and sensors to recognize facial expressions.

[0757] Server: Used to analyze collected data and adjust rewards and share information.

[0758] Job Information System: An online platform that provides job seekers with a viewable stress score.

[0759] software

[0760] Speech recognition algorithm: Analyzes collected voice data to determine whether abusive language is present.

[0761] Motion analysis algorithm: Analyzes collected motion data to determine whether violence is present or not.

[0762] Emotion engine: Software for analyzing emotional data and identifying stress levels.

[0763] Compensation Adjustment System: Automatically adjusts workers' compensation based on quantified information.

[0764] Information linkage system: Software for linking quantified information with a job information system.

[0765] Processing steps and functions

[0766] The server processes data, adjusts rewards, and shares information based on the following steps:

[0767] 1. Data collection: Workers wear wearable devices to collect voice, movement, and emotion data, such as microphones, accelerometers, and facial recognition sensors.

[0768] 2. Data Transfer: Collected data is transferred to the server in real time or periodically.

[0769] 3. Data analysis: The server uses a voice recognition algorithm to analyze the voice data and determine whether or not there is abusive language. It also uses a motion analysis algorithm to identify whether or not there is violence. An emotion engine analyzes the level of stress from the emotional data.

[0770] 4. Quantification: Calculate the worker's stress score based on the analyzed voice, movement, and emotional data. For example, a comprehensive score is generated that combines the frequency of verbal and physical abuse with emotional data.

[0771] 5. Compensation Adjustment: Based on the score, the compensation adjustment system will automatically adjust the worker's compensation, which is calculated according to certain criteria.

[0772] 6. Information sharing: The quantified information is transferred to the employment information system via the information sharing system, where it can be viewed by job seekers.

[0773] Specific examples

[0774] For example, suppose a factory worker wears a wearable device while working and collects the following data:

[0775] Voice data: phrases like "I'm not feeling too great today" and stressed voice tones are collected.

[0776] Movement data: Occasionally, certain movement patterns are recorded that indicate strain or fatigue.

[0777] Emotional data: Cameras and sensors detect workers' haggard expressions.

[0778] These data are input into the generative AI model as example prompt sentences as follows:

[0779] Prompt Sentence Examples

[0780] Imagine that Sato is wearing a wearable device while working, and generate the following data.

[0781] Audio data: "I'm not feeling too good today," "I'll try a little harder."

[0782] Movement data: movement patterns that indicate strain or fatigue

[0783] Emotion data: Haggard expression

[0784] These processes enable accurate assessment of worker stress, appropriate compensation adjustments, and visualization of the working environment.

[0785] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0786] Step 1:

[0787] The terminal collects voice, movement, and emotion data from the worker wearing the wearable device. Specifically, it uses a microphone to capture voice data, an accelerometer to measure movement data, and a facial recognition sensor to collect emotion data. This data is then transferred to a server in real time or periodically.

[0788] Step 2:

[0789] The server analyzes the received voice data. Using a voice recognition algorithm, it analyzes specific keywords and tone of voice in the data to determine whether it contains abusive language. The input is the voice data, and the output is a flag indicating whether abusive language is present. For example, if the word "idiot" is included, it is detected as abusive language.

[0790] Step 3:

[0791] The server analyzes the received motion data. Using a motion analysis algorithm, it analyzes specific motion patterns and determines whether a violent act has occurred. The input is the motion data, and the output is a flag indicating whether a violent act has occurred. For example, if a rapid change in acceleration is recorded, it is detected as a violent act.

[0792] Step 4:

[0793] The server uses an emotion engine to analyze the emotion data and calculate the stress level. The input is emotion data, and the output is a numerical value (stress score) that indicates the emotional state. Specifically, if haggard expressions are frequently detected, the stress score will be high.

[0794] Step 5:

[0795] The server integrates the results of the analysis of the voice data, movement data, and emotion data to calculate an overall stress score. Each piece of data is weighted to calculate the overall score. For example, the overall score is created by averaging the frequency of abusive language, the frequency of violent behavior, and emotion data indicating high stress. The inputs are the results of each analysis, and the output is the overall stress score.

[0796] Step 6:

[0797] The server automatically adjusts the worker's compensation based on the calculated overall stress score. It determines whether to increase or decrease compensation depending on the score and transfers the result to the payroll system. Specifically, if the overall score is high, bonuses and allowances are increased. The input is the overall stress score, and the output is the amount of compensation.

[0798] Step 7:

[0799] The server links the quantified stress score to the job information system. The information is made available to job seekers in a viewable format, allowing them to understand the working environment in advance. Specifically, the server displays the stress score for each workplace, helping job seekers select the appropriate workplace. The input is the overall stress score, and the output is the information displayed on the job information system.

[0800] 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.

[0801] 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.

[0802] 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.

[0803] [Third embodiment]

[0804] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0805] 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.

[0806] 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).

[0807] 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.

[0808] 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.

[0809] 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).

[0810] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0811] 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.

[0812] 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.

[0813] 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.

[0814] 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.

[0815] 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."

[0816] This invention is a system for visualizing stress in nursing care settings and reflecting it in compensation. This system is implemented through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, and providing information.

[0817] Program processing

[0818] 1. Data Collection

[0819] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using a microphone, and movement data is measured using an accelerometer. This data is periodically transferred to a server.

[0820] 2. Data Analysis

[0821] The server analyzes the received voice data using a voice recognition algorithm to identify the presence and frequency of abusive language, and uses a behavior analysis algorithm to detect violent behavior from the movement data. This increases the detection rate of abusive language and violence, and allows for more accurate data to be obtained.

[0822] 3. Quantification

[0823] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. This score is calculated based on a set of criteria and can be aggregated weekly, monthly, or yearly.

[0824] 4. Remuneration adjustment

[0825] The server automatically adjusts the compensation of each care worker based on the numerical stress score. Specifically, it calculates the amount of bonuses and allowances according to the score and reflects this in the payroll system.

[0826] 5. Information provision

[0827] The server then links the quantified stress scores to a job information system and makes them available to job seekers, allowing them to understand the working environment in advance and choose an appropriate workplace.

[0828] Specific examples

[0829] For example, let's say that caregiver Sato is wearing a wearable device and encounters the following situation.

[0830] Monday

[0831] While working, Sato was subjected to verbal abuse from users, such as "idiot" and "useless," twice. The accelerometer also recorded him being hit in the chest once.

[0832] Tuesday

[0833] Sato was subjected to one verbal abuse, "Stop it!", but no violence occurred.

[0834] The server receives this data and uses voice recognition and behavioral analysis algorithms to determine two cases of verbal abuse and one case of violence, then quantifies the results and reflects them in the weekly score.

[0835] Based on the stress score obtained in this way, the server automatically adjusts Sato's compensation and reflects it in the payroll system. At the same time, this quantified information is also linked to the job information system, allowing other job seekers to view and understand it in real time.

[0836] Through the above process, the present invention aims to reduce turnover by visualizing the stress of care workers and appropriately reflecting this in their compensation. This system will improve the working environment for care workers and support the recruitment and development of human resources throughout the care industry.

[0837] The processing flow will be explained below.

[0838] Step 1:

[0839] The user, a caregiver, puts on the wearable device and turns it on, and the device is automatically paired with the server.

[0840] Step 2:

[0841] The device collects real-time audio data through the wearable device's microphone, while also collecting body movement data using an accelerometer.

[0842] Step 3:

[0843] The device stores the collected voice data and motion data in a buffer at regular intervals (for example, every minute). The stored data is given a timestamp.

[0844] Step 4:

[0845] The terminal transmits all the data in the buffer to the server at once. If a transmission error occurs, the terminal performs a retransmission process.

[0846] Step 5:

[0847] The server performs pre-processing of the received audio data, which includes noise filtering and voice normalization.

[0848] Step 6:

[0849] The server runs a speech recognition algorithm to convert the preprocessed audio data into text, from which it uses natural language processing (NLP) to extract abusive keywords.

[0850] Step 7:

[0851] The server pre-processes the motion data, filtering the accelerometer data to remove inaccurate data.

[0852] Step 8:

[0853] The server runs a behavioral analysis algorithm to detect patterns of violence in the filtered motion data, based on sudden movements and impacts.

[0854] Step 9:

[0855] The server calculates a stress score for each caregiver based on the extracted abusive keywords and the detected patterns of violence. The score is calculated based on certain criteria.

[0856] Step 10:

[0857] The server aggregates weekly, monthly, and yearly stress scores and adjusts compensation accordingly, which is then linked to the payroll system.

[0858] Step 11:

[0859] The server connects the calculated stress score to the job information system and makes it publicly available for job seekers to view.

[0860] Step 12:

[0861] Users can browse job information pages and check the stress scores of each workplace to choose a workplace that suits them.

[0862] Example 1

[0863] 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."

[0864] In the field of nursing care, caregivers often experience high levels of stress on a daily basis, and are frequently subjected to verbal and physical abuse. This leads to a decline in staff motivation and a high turnover rate. However, there is no way to objectively grasp the actual level of stress felt by caregivers, and appropriate compensation adjustments are not being made. To solve this issue, there is a need for a system that can quantify the stress of caregivers and automatically adjust compensation based on this information.

[0865] 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.

[0866] In this invention, the server includes means for collecting voice data and movement data received by caregivers wearing wearable devices, means for analyzing the collected voice data and determining whether abusive language has been used, means for analyzing the collected movement data and determining whether violence has been used, means for quantifying the degree of abusive language and violence based on the analysis results, means for automatically adjusting the caregiver's compensation based on the quantified information, means for linking the quantified information to a recruitment system and providing it, means for collecting voice data using an acoustic sensor and movement data using a motion sensor, and means for aggregating the quantified information on a weekly, monthly, and yearly basis and automatically adjusting compensation based on the quantified information. This makes it possible to objectively quantify the stress of caregivers and adjust compensation appropriately.

[0867] A "wearable device" is an electronic device that is worn on the body and is used to collect voice data and movement data.

[0868] "Voice data" refers to the audio information received by caregivers, and is digital data collected using microphones and acoustic sensors.

[0869] "Motion data" refers to information that measures the physical movements and acceleration of care workers, and is digital data collected using acceleration sensors and motion sensors.

[0870] "Analysis" is the process of using collected data to extract specific information and make decisions based on that information.

[0871] "Abusive language" refers to insulting words or expressions directed at nursing care staff.

[0872] "Violence" refers to physical aggression or force used against care staff.

[0873] "Quantification" is the process of expressing analyzed data as quantitative numbers.

[0874] "Automatic compensation adjustment" is a system that automatically calculates and adjusts the salaries and bonuses of care workers based on quantified information.

[0875] The "recruitment system" refers to an online platform where job seekers can view information about nursing care workplaces and apply.

[0876] An "acoustic sensor" is a device that detects sound information and collects it as digital data.

[0877] A "motion sensor" is a device that detects body movements and collects them as digital data.

[0878] This invention is a system for visualizing stress in nursing care settings and reflecting it in compensation. This system is implemented through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, and providing information.

[0879] 1. Data Collection

[0880] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using an acoustic sensor (microphone), and movement data is measured using a motion sensor (accelerometer). This data is periodically transferred to a server.

[0881] 2. Data Transfer

[0882] The collected voice and movement data is transferred to a server in real time via Wi-Fi or Bluetooth, for example, every minute, split into packets and sent with error checking to prevent data loss.

[0883] 3. Data Analysis

[0884] The server converts the received audio data into text using the Google Cloud Speech-to-Text API and uses natural language processing (NLP) algorithms to identify the presence and frequency of abusive language. The movement data is analyzed using a custom behavioral analysis model using TensorFlow to detect physical violence. For example, certain keywords or high accelerations can be detected to identify abusive language or violence.

[0885] 4. Quantification

[0886] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. The quantification criteria are 10 points for each verbal abuse incident and 50 points for each physical abuse incident, and the data can be compiled weekly, monthly, or yearly.

[0887] 5. Remuneration adjustment

[0888] The server automatically adjusts the compensation of each care worker based on the quantified stress score. Specifically, it calculates the amount of bonuses and allowances according to the stress score and reflects this in the payroll system (e.g., SAP SuccessFactors). For example, it can be set up so that an additional bonus is paid to staff whose monthly score exceeds 300 points.

[0889] 6. Information provision

[0890] The server connects the quantified stress score to recruitment systems (e.g., Recruit and Indeed) so that job seekers can understand the working environment in advance. For example, if Mr. Sato has a high stress score, that information will be displayed on the recruitment information page, allowing other job seekers to use it as a reference when choosing a workplace.

[0891] Specific examples

[0892] For example, let's say that caregiver Sato is wearing a wearable device and encounters the following situation.

[0893] Monday

[0894] While working, Sato was twice subjected to abusive language from users, such as being called an idiot and being useless, and an accelerometer recorded him being hit in the chest once.

[0895] Tuesday

[0896] Sato was subjected to one verbal abuse of "Stop it!" but no violence occurred.

[0897] The server receives this data and, using a custom behavioral analysis model using the Google Cloud Speech-to-Text API and TensorFlow, determines two instances of verbal abuse and one instance of violence, converts the results into a numerical value, and reflects this in a weekly score. Based on the stress score obtained in this way, the server automatically adjusts Mr. Sato's compensation and reflects it in the payroll system (e.g., SAP SuccessFactors). At the same time, this numerical information is also linked to job information systems (e.g., Recruit and Indeed), allowing other job seekers to view and understand it in real time.

[0898] Example prompts to input to the generative AI model

[0899] Prompt: Using the data below, explain the specific steps for a system that calculates stress scores for caregivers and automatically adjusts compensation based on the results.

[0900] Monday: Sato was subjected to verbal abuse such as "idiot" and "useless" twice during work hours, and an accelerometer recorded him being hit in the chest once.

[0901] Tuesday: One verbal abuse of "Stop it!", no violence occurred.

[0902] The system proceeds as follows:

[0903] 1. Data collection (collection of voice data and movement data, transfer to server)

[0904] 2. Data analysis (voice recognition and behavior analysis)

[0905] 3. Quantification (calculating stress scores)

[0906] 4. Compensation adjustment (automatic compensation adjustment based on numerical data)

[0907] 5. Information provision (linking to recruitment information systems)

[0908] This invention aims to reduce turnover by visualizing stress in the nursing care field and appropriately reflecting it in compensation. This system will improve the working environment for nursing care workers and support the recruitment and development of human resources throughout the nursing care industry.

[0909] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0910] Step 1: Data collection

[0911] The terminal starts up the wearable device worn by the caregiver and starts collecting voice and movement data. Voice data is collected using an acoustic sensor, and movement data is measured using a motion sensor. For example, voice data (such as "idiot" or "useless") is collected here, and the acceleration sensor detects the action of hitting the chest. The input of this step is the voice and movement of the caregiver, and the output is the collected raw data.

[0912] Step 2: Data Transfer

[0913] The device transfers the collected voice and movement data to the server in real time via Wi-Fi or Bluetooth. For example, the voice and movement data is compressed every minute and sent to the server in packet format, minimizing data loss. The input of this step is the collected raw data, and the output is the raw data stored on the server.

[0914] Step 3: Data analysis

[0915] The server converts the received voice data into text using the Google Cloud Speech-to-Text API and analyzes the presence and frequency of abusive language using natural language processing (NLP) algorithms. It also analyzes the behavioral data using a custom behavioral analysis model using TensorFlow to identify violent behavior. For example, it detects keywords (e.g., "idiot" or "stop") or high accelerations in the converted text. The input of this step is the received raw data, and the output is the analysis results.

[0916] Step 4: Quantify

[0917] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. For example, points are assigned, such as 10 points for each verbal abuse and 50 points for each physical abuse, and the totals are used to calculate the stress score. The input for this step is the analysis results, and the output is the calculated stress score.

[0918] Step 5: Compensation Adjustment

[0919] The server automatically adjusts the compensation of each caregiver based on the quantified stress score. Specifically, it calculates the amount of bonuses and allowances according to the stress score and connects this to the payroll system. For example, if the monthly score exceeds 300 points, a bonus is added. The input to this step is the calculated stress score, and the output is the adjusted compensation information.

[0920] Step 6: Provide information

[0921] The server links the quantified stress score to the job information system and provides the data so that job seekers can view it. For example, a job seeker can check the stress environment of Mr. Sato's workplace on the job information page. The input of this step is the adjusted compensation information, and the output is the stress score reflected in the job information system.

[0922] Through these steps, this system will improve the working environment in nursing care settings, appropriately reflect the stress of nursing care staff in their compensation, and enable job seekers to understand the working environment in advance.

[0923] (Application example 1)

[0924] 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."

[0925] Currently, in many brick-and-mortar stores, there are many situations where staff members who serve customers feel stressed. When this stress increases, employee motivation decreases and it may even lead to an increase in turnover. Furthermore, in order to improve customer satisfaction, it is important to create an environment where staff can work comfortably, but there is a lack of means to achieve this. Furthermore, there is a need to introduce an appropriate compensation system to make staff stress visible and improve the working environment.

[0926] 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.

[0927] In this invention, the server includes means for collecting voice data and movement data received by workers wearing wearable devices, means for analyzing the collected voice data and determining whether or not there are words that cause stress, means for analyzing the collected movement data and determining whether or not there are movement events that indicate stress, means for quantifying the stress level based on the analysis results, means for adjusting the worker's remuneration based on the quantified information, means for linking the quantified information to a recruitment information system, and means for providing the quantified information to the worker. This makes it possible to grasp staff stress in real time, adjust remuneration based on that information, and provide a comfortable working environment.

[0928] A "wearable device" is a device that can be worn on the body and collects voice data and movement data in real time.

[0929] "Employee" refers to a worker who provides customer service in a physical store.

[0930] "Audio data" refers to audio signals collected using a microphone and recorded as digital data.

[0931] "Motion data" is digital data indicating body movements obtained using an acceleration sensor or other sensor device.

[0932] "Stressor words" refer to specific words or phrases that are important when assessing an employee's stress level, and are often negative comments from customers or words that indicate overwork.

[0933] "Movement event" refers to a specific physical movement detected using sensors, and may include movements that are indicative of stress.

[0934] The "stress level" is the estimated level of stress felt by the worker, calculated from the collected voice data and movement data.

[0935] "Remuneration" refers to the monetary compensation, including salary and bonuses, that an employee receives for their work.

[0936] A "job information system" is an online system that handles information on the labor market and provides job seekers with data on companies and jobs.

[0937] "Means of providing" refers to methods or systems for communicating collected data and calculated results to employees, such as notification systems or portal sites.

[0938] MODE FOR CARRYING OUT THE INVENTION

[0939] To realize this invention, store employees who serve customers must wear a wearable device. The wearable device is used to collect voice and motion data in real time. The collected data is sent to a server and analyzed using voice recognition and motion analysis algorithms.

[0940] A description of what the program does

[0941] Hardware and software configuration:

[0942] Hardware:

[0943] Wearable device: Equipped with a microphone and accelerometer to collect voice and movement data.

[0944] Accelerometer: A sensor that detects body movement.

[0945] Microphone: A device for collecting sound.

[0946] software:

[0947] Server: A computer system that performs analysis and stores data.

[0948] Speech recognition algorithms: convert collected voice data into text and detect specific stressful words.

[0949] Motion analysis algorithm: Analyzes acceleration data to detect motion events that indicate stress. As a concrete example, peak detection is performed using the Python library "scipy."

[0950] Data processing and calculations:

[0951] The voice data is converted into text by a speech recognition algorithm on the server, which checks whether certain keywords (e.g., "busy," "tired," "terrible," etc.) are included to determine the stress factor.

[0952] Movement data is analyzed based on the output of the acceleration sensor to detect movement events that are indicators of stress (e.g., large or unnatural movements).

[0953] These analysis results are quantified and the employee's stress score is calculated.

[0954] Information provision and compensation adjustment:

[0955] The quantified information is stored on a server and compiled periodically, weekly, monthly, and annually, and employee compensation is automatically adjusted based on the results.

[0956] This information will also be linked to the job information system so that job seekers can view it.

[0957] The collected and analyzed stress information is also provided to employees, helping them understand their own stress levels and consider ways to improve them.

[0958] Examples:

[0959] For example, if a customer service staff member at a brick-and-mortar store hears a comment from a customer that they are "busy" or "tired" and begins to feel stressed during their shift, the voice data is collected through a microphone. An accelerometer is used to simultaneously detect movement events indicative of stress, such as hand vibrations. This data is then sent to a server, where a speech recognition algorithm is used to analyze the voice data and detect the words "busy" and "tired." A movement analysis algorithm is then used to detect unnatural hand movements.

[0960] Example prompts to input to a generative AI model:

[0961] Please explain a system that uses voice and motion data to analyze store staff stress in real time, adjust compensation, and integrate job information. Please particularly focus on the specific use of voice recognition and motion analysis algorithms.

[0962] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0963] Step 1:

[0964] The terminal collects voice and movement data from the employee via a wearable device. The collected voice data is obtained through a microphone, and movement data is acquired from an acceleration sensor. The input is the voice and movement data collected in real time, which is then transferred to a server. The terminal periodically sends the data to the server.

[0965] Step 2:

[0966] The server converts the received voice data into text using a speech recognition algorithm. This algorithm uses the Python "speech_recognition" library. The input is voice data and the output is text data. The analyzed text is checked to see if it contains keywords that indicate stress.

[0967] Step 3:

[0968] The server analyzes the received motion data using a motion analysis algorithm. This algorithm uses Python's "scipy.signal.find_peaks". The input is the motion data and the output is the number of motion events. If the motion events exceed a certain threshold, it is considered an indicator of stress.

[0969] Step 4:

[0970] The server quantifies the stress level based on the results of the analysis of the voice data and movement data. The stress level is calculated based on the number of detected keywords and the number of movement events. The input is the analysis results of the voice data and movement data, and the output is a stress score.

[0971] Step 5:

[0972] The server adjusts the employee's compensation based on the calculated stress score. The compensation adjustment algorithm adjusts the base salary according to the stress score and adds a bonus if necessary. The input is the stress score, and the output is the adjusted compensation amount.

[0973] Step 6:

[0974] The server connects the quantified stress score to the job information system, allowing job seekers to understand the working environment in advance. The input is the stress score, and the output is the information published in the job information system.

[0975] Step 7:

[0976] The server provides employees with a numerical stress score, allowing them to understand their own stress level in real time and consider ways to improve it. The input is the stress score, and the output is feedback information provided to employees.

[0977] 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.

[0978] This invention aims to visualize stress in nursing care settings and reflect it in compensation, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more accurate stress assessment. This system runs through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, providing information, and recognizing emotions.

[0979] Program processing

[0980] 1. Data Collection

[0981] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using a microphone, and movement data is measured using an accelerometer. This data is periodically transferred to a server.

[0982] 2. Collecting Emotional Data

[0983] The terminal uses sensors built into the wearable device to collect emotional data such as facial expressions and body movements, which are then transmitted to the server along with voice and movement data.

[0984] 3. Data Analysis

[0985] The server analyzes the received voice data using a speech recognition algorithm to identify the presence and frequency of abusive language, the behavioral analysis algorithm to detect violent behavior, and the emotion engine to analyze the user's emotional state.

[0986] 4. Quantification

[0987] The server calculates a stress score for each caregiver based on the analyzed voice, movement, and emotion data. This score is calculated based on certain criteria and can be aggregated weekly, monthly, or yearly.

[0988] 5. Remuneration adjustment

[0989] The server automatically adjusts the compensation of each care worker based on the numerical stress score. Specifically, it calculates the amount of bonuses and allowances according to the score and reflects this in the payroll system.

[0990] 6. Information provision

[0991] The server then links the calculated stress score to a job information system and makes it available for job seekers to view, allowing them to understand the working environment in advance and choose an appropriate workplace.

[0992] Specific examples

[0993] For example, let's say that caregiver Yamada is wearing a wearable device and encounters the following situation.

[0994] Monday

[0995] While working, Yamada was subjected to verbal abuse from users twice, such as being called "stupid" and "useless." The acceleration sensor also recorded one instance of Yamada being tapped on the shoulder. Furthermore, the facial expression recognition function detected that Yamada was under stress.

[0996] Tuesday

[0997] Yamada was verbally abused once, with the man telling her to "Stop it!", but no violence occurred. However, analysis of the tone of his voice revealed that he was in a state of serious stress.

[0998] The server receives this data and uses a voice recognition algorithm, behavioral analysis algorithm, and emotion engine to determine two cases of verbal abuse, one case of violence, and a high level of emotional stress, then quantifies the results and reflects them in the weekly score.

[0999] Based on the stress score obtained in this way, the server automatically adjusts Yamada's compensation and reflects it in the payroll system. At the same time, the quantified information is also linked to the job information system, allowing other job seekers to view and understand it in real time.

[1000] Through the above process, the present invention aims to reduce turnover by more accurately visualizing the stress of care workers and appropriately reflecting it in their compensation. The introduction of an emotion engine will further improve the working environment for care workers and support the recruitment and development of human resources throughout the care industry.

[1001] The processing flow will be explained below.

[1002] Step 1:

[1003] The user, a caregiver, puts on the wearable device and turns it on, and the device is automatically paired with the server.

[1004] Step 2:

[1005] The device collects real-time voice and movement data through the wearable device's microphone and accelerometer, as well as emotional data using facial expression and other emotion recognition sensors.

[1006] Step 3:

[1007] The device stores the collected voice data, movement data, and emotion data in a buffer at regular intervals (for example, every minute). Each data is assigned a timestamp.

[1008] Step 4:

[1009] The terminal transmits all the data in the buffer to the server at once. If a transmission error occurs, the terminal performs a retransmission process.

[1010] Step 5:

[1011] The server pre-processes the received audio data, including noise filtering and voice normalization, and also pre-processes the collected emotion data to remove outliers.

[1012] Step 6:

[1013] The server runs a speech recognition algorithm to convert the preprocessed audio data into a string of text, from which natural language processing (NLP) is used to extract abusive keywords.

[1014] Step 7:

[1015] The server pre-processes the motion data, filtering the accelerometer data to remove inaccurate data.

[1016] Step 8:

[1017] The server runs a behavioral analysis algorithm to detect patterns of violence based on sudden movements and impacts from the accelerometer data.

[1018] Step 9:

[1019] The server uses an emotion engine to analyze the user's emotional state from the pre-processed emotion data, including analyzing voice tone, speech rate, volume, facial expressions, and body movements.

[1020] Step 10:

[1021] The server integrates information extracted from the voice data, movement data, and emotion data to calculate a stress score for each caregiver, which is quantified based on a set of criteria.

[1022] Step 11:

[1023] The server aggregates the calculated stress scores on a weekly, monthly, and yearly basis, and automatically adjusts the compensation of each care worker based on this.

[1024] Step 12:

[1025] The server connects the results of the compensation adjustment to the payroll system and reflects them in the salaries of each care worker.

[1026] Step 13:

[1027] The server links the quantified stress score to a job information system and makes it publicly available for job seekers to view.

[1028] Step 14:

[1029] Users can browse job information pages and check the stress scores and working environments of each workplace to choose a workplace that suits them.

[1030] As a concrete example, let's say that care worker Yamada wears a wearable device and receives two verbal abuses from a caregiver on Monday ("idiot" and "useless"), and receives one slap on the shoulder. If facial expression recognition detects that Yamada is stressed, the server analyzes this data and reflects a quantified stress score in the payroll system and job information system. This process allows Yamada to receive fair compensation, and job seekers can obtain accurate information about the work environment.

[1031] Example 2

[1032] 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."

[1033] There is a need for a method to visualize and accurately evaluate the stress caused by verbal abuse and violence experienced by caregivers in the nursing care field. Furthermore, there is a need to improve the working environment and reduce staff turnover by appropriately adjusting the compensation of caregivers based on this stress assessment. Furthermore, there is a need for a method to provide information that allows job seekers to understand the working environment in advance.

[1034] 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.

[1035] In this invention, the server includes: means for collecting voice data and movement data received by caregivers wearing wearable devices; means for analyzing the collected voice data to determine whether verbal abuse has occurred; means for analyzing the collected movement data to determine whether violence has occurred; means for collecting emotional data such as facial expressions and body movements to determine the emotional state of the caregiver; means for analyzing the collected emotional data to determine the emotional state of the caregiver; means for quantifying the voice data, movement data, and emotional data based on the analysis results and calculating the caregiver's stress score; means for adjusting the caregiver's compensation based on the quantified information; and means for linking the quantified information to a job information system. This allows for accurate visualization of caregiver stress and adjustment of compensation based on the assessment, thereby improving the working environment and reducing staff turnover. It also allows job seekers to understand the working environment in advance.

[1036] A "wearable device" is a portable device designed to be worn by caregivers and equipped with sensors, microphones, cameras, etc. to collect voice, movement, and emotional data.

[1037] "Audio data" refers to electronically recorded audio information such as conversations that occur during the work of caregivers and environmental sounds.

[1038] "Motion data" refers to electronically recorded data relating to the physical movements and behavior of caregivers, and is collected using an acceleration sensor or the like.

[1039] "Emotional data" refers to electronically recorded data that indicates the emotional state of a caregiver, as inferred from facial expressions, physical movements, heart rate, and electrodermal activity.

[1040] A "voice recognition algorithm" is a technology that analyzes collected voice data, converts it into text data, and detects specific keywords and abusive language.

[1041] "Behavioral analysis algorithms" are technologies that analyze collected behavioral data to detect violent acts and abnormal behavior.

[1042] The "emotion engine" is a technology that analyzes collected emotional data and determines the emotional state of caregivers.

[1043] The "stress score" is a numerical representation of the level of stress of a caregiver calculated based on the results of analyzing voice data, movement data, and emotion data.

[1044] "Remuneration adjustment means" refers to technology and systems for automatically calculating and changing the remuneration of care workers based on stress scores.

[1045] A "job information system" is a system that provides information about working conditions that is made publicly available for job seekers to view.

[1046] This invention aims to visualize stress in nursing care settings and reflect it in compensation, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more accurate stress assessment. This system runs through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, providing information, and recognizing emotions.

[1047] Hardware and Software Configuration

[1048] The terminal uses a wearable device (e.g., a smartwatch or fitness tracker) to collect voice, movement, and emotion data of caregivers in real time. Voice data is collected using the device's built-in microphone, movement data is measured using an acceleration sensor, and emotion data is collected using facial recognition sensors and heart rate sensors.

[1049] The server receives the collected data and performs the following analysis:

[1050] 1. Analysis of audio data

[1051] The server converts the audio data into text using a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API), then detects specific keywords and abusive language.

[1052] 2. Analysis of behavioral data

[1053] The server analyzes the motion data using a behavioral analysis algorithm to detect violent acts and abnormal behavior.

[1054] 3. Emotion Data Analysis

[1055] The server analyzes the emotion data using an emotion engine (e.g., Google Cloud Natural Language API) and determines the emotional state of the caregiver.

[1056] Specific examples

[1057] Let's say that caregiver Yamada wears a wearable device and encounters the following situation.

[1058] Monday

[1059] The device collects data on two occasions when Yamada was subjected to abusive language such as "idiot" and "useless" from users while working.

[1060] The device uses an accelerometer to record the tapping of Yamada's shoulder once.

[1061] The device's facial expression recognition function detects when Yamada is expressing stress.

[1062] The server analyzes this data and determines that there are two cases of verbal abuse, one case of violence, and a high level of stress, and quantifies this data.

[1063] Tuesday

[1064] The device collects data that it received one verbal abuse such as "Stop it!", but records that no violence occurred.

[1065] The device analyzes the voice tone and detects that Yamada is in a state of serious stress.

[1066] From this data, the server determines that one incident of abusive language is a high level of stress and converts it into a numerical value.

[1067] Examples of prompt statements

[1068] An example of a prompt to be input to the generative AI model is as follows:

[1069] Please explain the system that calculates the stress score of caregivers based on the voice data, movement data, and emotion data collected from the caregivers' wearable devices, and automatically adjusts compensation based on that score. The voice data is used to check the presence and frequency of abusive language, and the movement data is used to identify violent acts. The emotion data is used to analyze the stress level using an emotion engine.

[1070] This system will accurately visualize the stress levels of care workers and adjust compensation based on the results, which will improve the working environment and reduce staff turnover. It will also enable job seekers to understand the working environment in advance.

[1071] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1072] Step 1: Data collection

[1073] The terminal collects voice data, movement data, and emotion data of the caregiver wearing the wearable device.

[1074] Input: Voice, movement, and facial expression generated by caregivers

[1075] Specific behavior:

[1076] The device uses a built-in microphone to collect audio data, such as the content of conversations and background sounds.

[1077] The device uses an integrated accelerometer to collect motion data, such as the number of steps taken by the caregiver, posture, and sudden movements.

[1078] The device collects emotional data using built-in facial recognition and heart rate sensors, for example, recording the caregiver's facial expressions and heart rate.

[1079] Output: Collected voice data, movement data, and emotion data

[1080] Step 2: Data Transfer

[1081] The terminal periodically transfers the collected data to the server.

[1082] Input: Collected voice data, movement data, emotion data

[1083] Specific behavior:

[1084] The terminal aggregates the collected data at regular intervals, encrypts it, and sends it to the server.

[1085] Output: Voice data, movement data, and emotion data transferred to the server

[1086] Step 3: Analyzing the audio data

[1087] The server analyzes the received voice data and determines whether or not abusive language is used.

[1088] Input: Audio data sent from the device

[1089] Specific behavior:

[1090] The server converts the voice data into text using a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API).

[1091] The server detects specific keywords and abusive language in the text data.

[1092] Output: Whether or not abusive language was detected and its frequency

[1093] Step 4: Analyzing the behavioral data

[1094] The server analyzes the received motion data and determines whether or not violence has occurred.

[1095] Input: Operation data sent from the device

[1096] Specific behavior:

[1097] The server uses a behavioral analysis algorithm to detect violent acts and abnormal behavior from the motion data.

[1098] For example, violent acts can be identified from abnormal acceleration values ​​or specific movement patterns.

[1099] Output: Whether or not violent acts were detected and their frequency

[1100] Step 5: Analyze the sentiment data

[1101] The server analyzes the received emotion data and determines the emotional state of the caregiver.

[1102] Input: Emotion data sent from the device

[1103] Specific behavior:

[1104] The server uses an emotion engine (e.g., Google Cloud Natural Language API) to classify the emotional state based on data from facial expression recognition sensors and heart rate sensors.

[1105] For example, it can recognize emotions such as smiles, anxiety, and anger from facial expressions, and estimate the level of stress from fluctuations in heart rate.

[1106] Output: Determined emotional state

[1107] Step 6: Calculate your stress score

[1108] The server calculates the stress score of the caregiver based on the analysis results.

[1109] Input: Analyzed voice data, movement data, emotion data

[1110] Specific behavior:

[1111] The server combines the information obtained from each data point and calculates a stress score, taking into account, for example, the number of abusive verbal and violent acts and changes in emotional state.

[1112] Output: Calculated stress score

[1113] Step 7: Compensation Adjustment

[1114] The server adjusts the caregiver's compensation based on the calculated stress score.

[1115] Input: Calculated stress score

[1116] Specific behavior:

[1117] The server calculates the amount of bonus or allowance based on the stress score.

[1118] The calculation results are linked to the payroll system and reflected in monthly pay slips.

[1119] Output: Adjusted reward information

[1120] Step 8: Provide information

[1121] The server links the stress score to a job information system and makes it publicly available for job seekers to view.

[1122] Input: Calculated stress score

[1123] Specific behavior:

[1124] The server uploads the stress score to the job information system and makes it public.

[1125] This allows job seekers to understand the working environment in advance.

[1126] Output: Published stress scores and work environment information

[1127] (Application example 2)

[1128] 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."

[1129] Conventional systems lacked a method for accurately assessing the stress levels of workers in nursing care and factory settings and directly reflecting the results in their compensation. In particular, stress assessments that took emotional data into account were not performed, creating challenges in improving the working environment and increasing worker motivation. Furthermore, there were also insufficient methods for visualizing workers' stress levels in detail and providing information that would allow job seekers to know the working environment in advance.

[1130] 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.

[1131] In this invention, the server includes means for collecting voice data and motion data received by workers wearing wearable devices, means for analyzing the collected voice data and determining whether abusive language has been used, means for analyzing the collected motion data and determining whether violence has been used, means for analyzing the emotion data using an emotion engine and identifying the level of stress, means for quantifying the level of abusive language, violence, and emotion data based on the analysis results, means for adjusting worker compensation based on the quantified information, and means for linking the quantified information to a job information system. This makes it possible to accurately assess worker stress and adjust compensation based on the assessment, as well as enable job seekers to understand the working environment in advance.

[1132] A "wearable device" is a device that can be worn by workers and collects voice data, movement data, emotional data, etc.

[1133] "Audio data" is digital data that records the words spoken by workers and the sounds of their surroundings.

[1134] "Movement data" is digital data collected by sensors on specific physical movements of workers.

[1135] "Abusive language" refers to words that have an insulting or negative connotation towards others.

[1136] "Violence" refers to the act of physically aggressing or inflicting harm on another person.

[1137] An "emotion engine" is software or algorithms that analyze the emotional state of workers based on collected data.

[1138] "Stress level" is a numerical representation of a worker's mental and emotional burden.

[1139] "Quantification" is the process of converting qualitative data into quantitative data that can be treated as such.

[1140] "Remuneration" refers to the monetary compensation received by workers, such as salary, bonuses, and allowances.

[1141] A "job information system" is an online platform where companies and facilities provide job information to job seekers.

[1142] The system for realizing this invention is composed of a wearable device, a server, and a job information system. The processing content of each step, the hardware and software used, and specific examples are explained in detail below.

[1143] Hardware and Software

[1144] Hardware

[1145] Wearable devices: Worn by workers, these devices include microphones to collect voice data, accelerometers to collect movement data, and cameras and sensors to recognize facial expressions.

[1146] Server: Used to analyze collected data and adjust rewards and share information.

[1147] Job Information System: An online platform that provides job seekers with a viewable stress score.

[1148] software

[1149] Speech recognition algorithm: Analyzes collected voice data to determine whether abusive language is present.

[1150] Motion analysis algorithm: Analyzes collected motion data to determine whether violence is present or not.

[1151] Emotion engine: Software for analyzing emotional data and identifying stress levels.

[1152] Compensation Adjustment System: Automatically adjusts workers' compensation based on quantified information.

[1153] Information linkage system: Software for linking quantified information with a job information system.

[1154] Processing steps and functions

[1155] The server processes data, adjusts rewards, and shares information based on the following steps:

[1156] 1. Data collection: Workers wear wearable devices to collect voice, movement, and emotion data, such as microphones, accelerometers, and facial recognition sensors.

[1157] 2. Data Transfer: Collected data is transferred to the server in real time or periodically.

[1158] 3. Data analysis: The server uses a voice recognition algorithm to analyze the voice data and determine whether or not there is abusive language. It also uses a motion analysis algorithm to identify whether or not there is violence. An emotion engine analyzes the level of stress from the emotional data.

[1159] 4. Quantification: Calculate the worker's stress score based on the analyzed voice, movement, and emotional data. For example, a comprehensive score is generated that combines the frequency of verbal and physical abuse with emotional data.

[1160] 5. Compensation Adjustment: Based on the score, the compensation adjustment system will automatically adjust the worker's compensation, which is calculated according to certain criteria.

[1161] 6. Information sharing: The quantified information is transferred to the employment information system via the information sharing system, where it can be viewed by job seekers.

[1162] Specific examples

[1163] For example, suppose a factory worker wears a wearable device while working and collects the following data:

[1164] Voice data: phrases like "I'm not feeling too great today" and stressed voice tones are collected.

[1165] Movement data: Occasionally, certain movement patterns are recorded that indicate strain or fatigue.

[1166] Emotional data: Cameras and sensors detect workers' haggard expressions.

[1167] These data are input into the generative AI model as example prompt sentences as follows:

[1168] Prompt Sentence Examples

[1169] Imagine that Sato is wearing a wearable device while working, and generate the following data.

[1170] Audio data: "I'm not feeling too good today," "I'll try a little harder."

[1171] Movement data: movement patterns that indicate strain or fatigue

[1172] Emotion data: Haggard expression

[1173] These processes enable accurate assessment of worker stress, appropriate compensation adjustments, and visualization of the working environment.

[1174] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1175] Step 1:

[1176] The terminal collects voice, movement, and emotion data from the worker wearing the wearable device. Specifically, it uses a microphone to capture voice data, an accelerometer to measure movement data, and a facial recognition sensor to collect emotion data. This data is then transferred to a server in real time or periodically.

[1177] Step 2:

[1178] The server analyzes the received voice data. Using a voice recognition algorithm, it analyzes specific keywords and tone of voice in the data to determine whether it contains abusive language. The input is the voice data, and the output is a flag indicating whether abusive language is present. For example, if the word "idiot" is included, it is detected as abusive language.

[1179] Step 3:

[1180] The server analyzes the received motion data. Using a motion analysis algorithm, it analyzes specific motion patterns and determines whether a violent act has occurred. The input is the motion data, and the output is a flag indicating whether a violent act has occurred. For example, if a rapid change in acceleration is recorded, it is detected as a violent act.

[1181] Step 4:

[1182] The server uses an emotion engine to analyze the emotion data and calculate the stress level. The input is emotion data, and the output is a numerical value (stress score) that indicates the emotional state. Specifically, if haggard expressions are frequently detected, the stress score will be high.

[1183] Step 5:

[1184] The server integrates the results of the analysis of the voice data, movement data, and emotion data to calculate an overall stress score. Each piece of data is weighted to calculate the overall score. For example, the overall score is created by averaging the frequency of abusive language, the frequency of violent behavior, and emotion data indicating high stress. The inputs are the results of each analysis, and the output is the overall stress score.

[1185] Step 6:

[1186] The server automatically adjusts the worker's compensation based on the calculated overall stress score. It determines whether to increase or decrease compensation depending on the score and transfers the result to the payroll system. Specifically, if the overall score is high, bonuses and allowances are increased. The input is the overall stress score, and the output is the amount of compensation.

[1187] Step 7:

[1188] The server links the quantified stress score to the job information system. The information is made available to job seekers in a viewable format, allowing them to understand the working environment in advance. Specifically, the server displays the stress score for each workplace, helping job seekers select the appropriate workplace. The input is the overall stress score, and the output is the information displayed on the job information system.

[1189] 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.

[1190] 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.

[1191] 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.

[1192] [Fourth embodiment]

[1193] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1194] 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.

[1195] 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).

[1196] 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.

[1197] 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.

[1198] 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).

[1199] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1200] 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.

[1201] 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.

[1202] 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.

[1203] 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.

[1204] 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.

[1205] 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."

[1206] This invention is a system for visualizing stress in nursing care settings and reflecting it in compensation. This system is implemented through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, and providing information.

[1207] Program processing

[1208] 1. Data Collection

[1209] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using a microphone, and movement data is measured using an accelerometer. This data is periodically transferred to a server.

[1210] 2. Data Analysis

[1211] The server analyzes the received voice data using a voice recognition algorithm to identify the presence and frequency of abusive language, and uses a behavior analysis algorithm to detect violent behavior from the movement data. This increases the detection rate of abusive language and violence, and allows for more accurate data to be obtained.

[1212] 3. Quantification

[1213] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. This score is calculated based on a set of criteria and can be aggregated weekly, monthly, or yearly.

[1214] 4. Remuneration adjustment

[1215] The server automatically adjusts the compensation of each care worker based on the numerical stress score. Specifically, it calculates the amount of bonuses and allowances according to the score and reflects this in the payroll system.

[1216] 5. Information provision

[1217] The server then links the quantified stress scores to a job information system and makes them available to job seekers, allowing them to understand the working environment in advance and choose an appropriate workplace.

[1218] Specific examples

[1219] For example, let's say that caregiver Sato is wearing a wearable device and encounters the following situation.

[1220] Monday

[1221] While working, Sato was subjected to verbal abuse from users, such as "idiot" and "useless," twice. The accelerometer also recorded him being hit in the chest once.

[1222] Tuesday

[1223] Sato was subjected to one verbal abuse, "Stop it!", but no violence occurred.

[1224] The server receives this data and uses voice recognition and behavioral analysis algorithms to determine two cases of verbal abuse and one case of violence, then quantifies the results and reflects them in the weekly score.

[1225] Based on the stress score obtained in this way, the server automatically adjusts Sato's compensation and reflects it in the payroll system. At the same time, this quantified information is also linked to the job information system, allowing other job seekers to view and understand it in real time.

[1226] Through the above process, the present invention aims to reduce turnover by visualizing the stress of care workers and appropriately reflecting this in their compensation. This system will improve the working environment for care workers and support the recruitment and development of human resources throughout the care industry.

[1227] The processing flow will be explained below.

[1228] Step 1:

[1229] The user, a caregiver, puts on the wearable device and turns it on, and the device is automatically paired with the server.

[1230] Step 2:

[1231] The device collects real-time audio data through the wearable device's microphone, while also collecting body movement data using an accelerometer.

[1232] Step 3:

[1233] The device stores the collected voice data and motion data in a buffer at regular intervals (for example, every minute). The stored data is given a timestamp.

[1234] Step 4:

[1235] The terminal transmits all the data in the buffer to the server at once. If a transmission error occurs, the terminal performs a retransmission process.

[1236] Step 5:

[1237] The server performs pre-processing of the received audio data, which includes noise filtering and voice normalization.

[1238] Step 6:

[1239] The server runs a speech recognition algorithm to convert the preprocessed audio data into text, from which it uses natural language processing (NLP) to extract abusive keywords.

[1240] Step 7:

[1241] The server pre-processes the motion data, filtering the accelerometer data to remove inaccurate data.

[1242] Step 8:

[1243] The server runs a behavioral analysis algorithm to detect patterns of violence in the filtered motion data, based on sudden movements and impacts.

[1244] Step 9:

[1245] The server calculates a stress score for each caregiver based on the extracted abusive keywords and the detected patterns of violence. The score is calculated based on certain criteria.

[1246] Step 10:

[1247] The server aggregates weekly, monthly, and yearly stress scores and adjusts compensation accordingly, which is then linked to the payroll system.

[1248] Step 11:

[1249] The server connects the calculated stress score to the job information system and makes it publicly available for job seekers to view.

[1250] Step 12:

[1251] Users can browse job information pages and check the stress scores of each workplace to choose a workplace that suits them.

[1252] Example 1

[1253] 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."

[1254] In the field of nursing care, caregivers often experience high levels of stress on a daily basis, and are frequently subjected to verbal and physical abuse. This leads to a decline in staff motivation and a high turnover rate. However, there is no way to objectively grasp the actual level of stress felt by caregivers, and appropriate compensation adjustments are not being made. To solve this issue, there is a need for a system that can quantify the stress of caregivers and automatically adjust compensation based on this information.

[1255] 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.

[1256] In this invention, the server includes means for collecting voice data and movement data received by caregivers wearing wearable devices, means for analyzing the collected voice data and determining whether abusive language has been used, means for analyzing the collected movement data and determining whether violence has been used, means for quantifying the degree of abusive language and violence based on the analysis results, means for automatically adjusting the caregiver's compensation based on the quantified information, means for linking the quantified information to a recruitment system and providing it, means for collecting voice data using an acoustic sensor and movement data using a motion sensor, and means for aggregating the quantified information on a weekly, monthly, and yearly basis and automatically adjusting compensation based on the quantified information. This makes it possible to objectively quantify the stress of caregivers and adjust compensation appropriately.

[1257] A "wearable device" is an electronic device that is worn on the body and is used to collect voice data and movement data.

[1258] "Voice data" refers to the audio information received by caregivers, and is digital data collected using microphones and acoustic sensors.

[1259] "Motion data" refers to information that measures the physical movements and acceleration of care workers, and is digital data collected using acceleration sensors and motion sensors.

[1260] "Analysis" is the process of using collected data to extract specific information and make decisions based on that information.

[1261] "Abusive language" refers to insulting words or expressions directed at nursing care staff.

[1262] "Violence" refers to physical aggression or force used against care staff.

[1263] "Quantification" is the process of expressing analyzed data as quantitative numbers.

[1264] "Automatic compensation adjustment" is a system that automatically calculates and adjusts the salaries and bonuses of care workers based on quantified information.

[1265] The "recruitment system" refers to an online platform where job seekers can view information about nursing care workplaces and apply.

[1266] An "acoustic sensor" is a device that detects sound information and collects it as digital data.

[1267] A "motion sensor" is a device that detects body movements and collects them as digital data.

[1268] This invention is a system for visualizing stress in nursing care settings and reflecting it in compensation. This system is implemented through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, and providing information.

[1269] 1. Data Collection

[1270] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using an acoustic sensor (microphone), and movement data is measured using a motion sensor (accelerometer). This data is periodically transferred to a server.

[1271] 2. Data Transfer

[1272] The collected voice and movement data is transferred to a server in real time via Wi-Fi or Bluetooth, for example, every minute, split into packets and sent with error checking to prevent data loss.

[1273] 3. Data Analysis

[1274] The server converts the received audio data into text using the Google Cloud Speech-to-Text API and uses natural language processing (NLP) algorithms to identify the presence and frequency of abusive language. The movement data is analyzed using a custom behavioral analysis model using TensorFlow to detect physical violence. For example, certain keywords or high accelerations can be detected to identify abusive language or violence.

[1275] 4. Quantification

[1276] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. The quantification criteria are 10 points for each verbal abuse incident and 50 points for each physical abuse incident, and the data can be compiled weekly, monthly, or yearly.

[1277] 5. Remuneration adjustment

[1278] The server automatically adjusts the compensation of each care worker based on the quantified stress score. Specifically, it calculates the amount of bonuses and allowances according to the stress score and reflects this in the payroll system (e.g., SAP SuccessFactors). For example, it can be set up so that an additional bonus is paid to staff whose monthly score exceeds 300 points.

[1279] 6. Information provision

[1280] The server connects the quantified stress score to recruitment systems (e.g., Recruit and Indeed) so that job seekers can understand the working environment in advance. For example, if Mr. Sato has a high stress score, that information will be displayed on the recruitment information page, allowing other job seekers to use it as a reference when choosing a workplace.

[1281] Specific examples

[1282] For example, let's say that caregiver Sato is wearing a wearable device and encounters the following situation.

[1283] Monday

[1284] While working, Sato was twice subjected to abusive language from users, such as being called an idiot and being useless, and an accelerometer recorded him being hit in the chest once.

[1285] Tuesday

[1286] Sato was subjected to one verbal abuse of "Stop it!" but no violence occurred.

[1287] The server receives this data and, using a custom behavioral analysis model using the Google Cloud Speech-to-Text API and TensorFlow, determines two instances of verbal abuse and one instance of violence, converts the results into a numerical value, and reflects this in a weekly score. Based on the stress score obtained in this way, the server automatically adjusts Mr. Sato's compensation and reflects it in the payroll system (e.g., SAP SuccessFactors). At the same time, this numerical information is also linked to job information systems (e.g., Recruit and Indeed), allowing other job seekers to view and understand it in real time.

[1288] Example prompts to input to the generative AI model

[1289] Prompt: Using the data below, explain the specific steps for a system that calculates stress scores for caregivers and automatically adjusts compensation based on the results.

[1290] Monday: Sato was subjected to verbal abuse such as "idiot" and "useless" twice during work hours, and an accelerometer recorded him being hit in the chest once.

[1291] Tuesday: One verbal abuse of "Stop it!", no violence occurred.

[1292] The system proceeds as follows:

[1293] 1. Data collection (collection of voice data and movement data, transfer to server)

[1294] 2. Data analysis (voice recognition and behavior analysis)

[1295] 3. Quantification (calculating stress scores)

[1296] 4. Compensation adjustment (automatic compensation adjustment based on numerical data)

[1297] 5. Information provision (linking to recruitment information systems)

[1298] This invention aims to reduce turnover by visualizing stress in the nursing care field and appropriately reflecting it in compensation. This system will improve the working environment for nursing care workers and support the recruitment and development of human resources throughout the nursing care industry.

[1299] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1300] Step 1: Data collection

[1301] The terminal starts up the wearable device worn by the caregiver and starts collecting voice and movement data. Voice data is collected using an acoustic sensor, and movement data is measured using a motion sensor. For example, voice data (such as "idiot" or "useless") is collected here, and the acceleration sensor detects the action of hitting the chest. The input of this step is the voice and movement of the caregiver, and the output is the collected raw data.

[1302] Step 2: Data Transfer

[1303] The device transfers the collected voice and movement data to the server in real time via Wi-Fi or Bluetooth. For example, the voice and movement data is compressed every minute and sent to the server in packet format, minimizing data loss. The input of this step is the collected raw data, and the output is the raw data stored on the server.

[1304] Step 3: Data analysis

[1305] The server converts the received voice data into text using the Google Cloud Speech-to-Text API and analyzes the presence and frequency of abusive language using natural language processing (NLP) algorithms. It also analyzes the behavioral data using a custom behavioral analysis model using TensorFlow to identify violent behavior. For example, it detects keywords (e.g., "idiot" or "stop") or high accelerations in the converted text. The input of this step is the received raw data, and the output is the analysis results.

[1306] Step 4: Quantify

[1307] The server quantifies the analyzed verbal and physical abuse data and calculates a stress score for each caregiver. For example, points are assigned, such as 10 points for each verbal abuse and 50 points for each physical abuse, and the totals are used to calculate the stress score. The input for this step is the analysis results, and the output is the calculated stress score.

[1308] Step 5: Compensation Adjustment

[1309] The server automatically adjusts the compensation of each caregiver based on the quantified stress score. Specifically, it calculates the amount of bonuses and allowances according to the stress score and connects this to the payroll system. For example, if the monthly score exceeds 300 points, a bonus is added. The input to this step is the calculated stress score, and the output is the adjusted compensation information.

[1310] Step 6: Provide information

[1311] The server links the quantified stress score to the job information system and provides the data so that job seekers can view it. For example, a job seeker can check the stress environment of Mr. Sato's workplace on the job information page. The input of this step is the adjusted compensation information, and the output is the stress score reflected in the job information system.

[1312] Through these steps, this system will improve the working environment in nursing care settings, appropriately reflect the stress of nursing care staff in their compensation, and enable job seekers to understand the working environment in advance.

[1313] (Application example 1)

[1314] 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."

[1315] Currently, in many brick-and-mortar stores, there are many situations where staff members who serve customers feel stressed. When this stress increases, employee motivation decreases and it may even lead to an increase in turnover. Furthermore, in order to improve customer satisfaction, it is important to create an environment where staff can work comfortably, but there is a lack of means to achieve this. Furthermore, there is a need to introduce an appropriate compensation system to make staff stress visible and improve the working environment.

[1316] 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.

[1317] In this invention, the server includes means for collecting voice data and movement data received by workers wearing wearable devices, means for analyzing the collected voice data and determining whether or not there are words that cause stress, means for analyzing the collected movement data and determining whether or not there are movement events that indicate stress, means for quantifying the stress level based on the analysis results, means for adjusting the worker's remuneration based on the quantified information, means for linking the quantified information to a recruitment information system, and means for providing the quantified information to the worker. This makes it possible to grasp staff stress in real time, adjust remuneration based on that information, and provide a comfortable working environment.

[1318] A "wearable device" is a device that can be worn on the body and collects voice data and movement data in real time.

[1319] "Employee" refers to a worker who provides customer service in a physical store.

[1320] "Audio data" refers to audio signals collected using a microphone and recorded as digital data.

[1321] "Motion data" is digital data indicating body movements obtained using an acceleration sensor or other sensor device.

[1322] "Stressor words" refer to specific words or phrases that are important when assessing an employee's stress level, and are often negative comments from customers or words that indicate overwork.

[1323] "Movement event" refers to a specific physical movement detected using sensors, and may include movements that are indicative of stress.

[1324] The "stress level" is the estimated level of stress felt by the worker, calculated from the collected voice data and movement data.

[1325] "Remuneration" refers to the monetary compensation, including salary and bonuses, that an employee receives for their work.

[1326] A "job information system" is an online system that handles information on the labor market and provides job seekers with data on companies and jobs.

[1327] "Means of providing" refers to methods or systems for communicating collected data and calculated results to employees, such as notification systems or portal sites.

[1328] MODE FOR CARRYING OUT THE INVENTION

[1329] To realize this invention, store employees who serve customers must wear a wearable device. The wearable device is used to collect voice and motion data in real time. The collected data is sent to a server and analyzed using voice recognition and motion analysis algorithms.

[1330] A description of what the program does

[1331] Hardware and software configuration:

[1332] Hardware:

[1333] Wearable device: Equipped with a microphone and accelerometer to collect voice and movement data.

[1334] Accelerometer: A sensor that detects body movement.

[1335] Microphone: A device for collecting sound.

[1336] software:

[1337] Server: A computer system that performs analysis and stores data.

[1338] Speech recognition algorithms: convert collected voice data into text and detect specific stressful words.

[1339] Motion analysis algorithm: Analyzes acceleration data to detect motion events that indicate stress. As a concrete example, peak detection is performed using the Python library "scipy."

[1340] Data processing and calculations:

[1341] The voice data is converted into text by a speech recognition algorithm on the server, which checks whether certain keywords (e.g., "busy," "tired," "terrible," etc.) are included to determine the stress factor.

[1342] Movement data is analyzed based on the output of the acceleration sensor to detect movement events that are indicators of stress (e.g., large or unnatural movements).

[1343] These analysis results are quantified and the employee's stress score is calculated.

[1344] Information provision and compensation adjustment:

[1345] The quantified information is stored on a server and compiled periodically, weekly, monthly, and annually, and employee compensation is automatically adjusted based on the results.

[1346] This information will also be linked to the job information system so that job seekers can view it.

[1347] The collected and analyzed stress information is also provided to employees, helping them understand their own stress levels and consider ways to improve them.

[1348] Examples:

[1349] For example, if a customer service staff member at a brick-and-mortar store hears a comment from a customer that they are "busy" or "tired" and begins to feel stressed during their shift, the voice data is collected through a microphone. An accelerometer is used to simultaneously detect movement events indicative of stress, such as hand vibrations. This data is then sent to a server, where a speech recognition algorithm is used to analyze the voice data and detect the words "busy" and "tired." A movement analysis algorithm is then used to detect unnatural hand movements.

[1350] Example prompts to input to a generative AI model:

[1351] Please explain a system that uses voice and motion data to analyze store staff stress in real time, adjust compensation, and integrate job information. Please particularly focus on the specific use of voice recognition and motion analysis algorithms.

[1352] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1353] Step 1:

[1354] The terminal collects voice and movement data from the employee via a wearable device. The collected voice data is obtained through a microphone, and movement data is acquired from an acceleration sensor. The input is the voice and movement data collected in real time, which is then transferred to a server. The terminal periodically sends the data to the server.

[1355] Step 2:

[1356] The server converts the received voice data into text using a speech recognition algorithm. This algorithm uses the Python "speech_recognition" library. The input is voice data and the output is text data. The analyzed text is checked to see if it contains keywords that indicate stress.

[1357] Step 3:

[1358] The server analyzes the received motion data using a motion analysis algorithm. This algorithm uses Python's "scipy.signal.find_peaks". The input is the motion data and the output is the number of motion events. If the motion events exceed a certain threshold, it is considered an indicator of stress.

[1359] Step 4:

[1360] The server quantifies the stress level based on the results of the analysis of the voice data and movement data. The stress level is calculated based on the number of detected keywords and the number of movement events. The input is the analysis results of the voice data and movement data, and the output is a stress score.

[1361] Step 5:

[1362] The server adjusts the employee's compensation based on the calculated stress score. The compensation adjustment algorithm adjusts the base salary according to the stress score and adds a bonus if necessary. The input is the stress score, and the output is the adjusted compensation amount.

[1363] Step 6:

[1364] The server connects the quantified stress score to the job information system, allowing job seekers to understand the working environment in advance. The input is the stress score, and the output is the information published in the job information system.

[1365] Step 7:

[1366] The server provides employees with a numerical stress score, allowing them to understand their own stress level in real time and consider ways to improve it. The input is the stress score, and the output is feedback information provided to employees.

[1367] 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.

[1368] This invention aims to visualize stress in nursing care settings and reflect it in compensation, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more accurate stress assessment. This system runs through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, providing information, and recognizing emotions.

[1369] Program processing

[1370] 1. Data Collection

[1371] The terminal collects real-time voice and movement data through a wearable device worn by the caregiver. Voice data is collected using a microphone, and movement data is measured using an accelerometer. This data is periodically transferred to a server.

[1372] 2. Collecting Emotional Data

[1373] The terminal uses sensors built into the wearable device to collect emotional data such as facial expressions and body movements, which are then transmitted to the server along with voice and movement data.

[1374] 3. Data Analysis

[1375] The server analyzes the received voice data using a speech recognition algorithm to identify the presence and frequency of abusive language, the behavioral analysis algorithm to detect violent behavior, and the emotion engine to analyze the user's emotional state.

[1376] 4. Quantification

[1377] The server calculates a stress score for each caregiver based on the analyzed voice, movement, and emotion data. This score is calculated based on certain criteria and can be aggregated weekly, monthly, or yearly.

[1378] 5. Remuneration adjustment

[1379] The server automatically adjusts the compensation of each care worker based on the numerical stress score. Specifically, it calculates the amount of bonuses and allowances according to the score and reflects this in the payroll system.

[1380] 6. Information provision

[1381] The server then links the calculated stress score to a job information system and makes it available for job seekers to view, allowing them to understand the working environment in advance and choose an appropriate workplace.

[1382] Specific examples

[1383] For example, let's say that caregiver Yamada is wearing a wearable device and encounters the following situation.

[1384] Monday

[1385] While working, Yamada was subjected to verbal abuse from users twice, such as being called "stupid" and "useless." The acceleration sensor also recorded one instance of Yamada being tapped on the shoulder. Furthermore, the facial expression recognition function detected that Yamada was under stress.

[1386] Tuesday

[1387] Yamada was verbally abused once, with the man telling her to "Stop it!", but no violence occurred. However, analysis of the tone of his voice revealed that he was in a state of serious stress.

[1388] The server receives this data and uses a voice recognition algorithm, behavioral analysis algorithm, and emotion engine to determine two cases of verbal abuse, one case of violence, and a high level of emotional stress, then quantifies the results and reflects them in the weekly score.

[1389] Based on the stress score obtained in this way, the server automatically adjusts Yamada's compensation and reflects it in the payroll system. At the same time, the quantified information is also linked to the job information system, allowing other job seekers to view and understand it in real time.

[1390] Through the above process, the present invention aims to reduce turnover by more accurately visualizing the stress of care workers and appropriately reflecting it in their compensation. The introduction of an emotion engine will further improve the working environment for care workers and support the recruitment and development of human resources throughout the care industry.

[1391] The processing flow will be explained below.

[1392] Step 1:

[1393] The user, a caregiver, puts on the wearable device and turns it on, and the device is automatically paired with the server.

[1394] Step 2:

[1395] The device collects real-time voice and movement data through the wearable device's microphone and accelerometer, as well as emotional data using facial expression and other emotion recognition sensors.

[1396] Step 3:

[1397] The device stores the collected voice data, movement data, and emotion data in a buffer at regular intervals (for example, every minute). Each data is assigned a timestamp.

[1398] Step 4:

[1399] The terminal transmits all the data in the buffer to the server at once. If a transmission error occurs, the terminal performs a retransmission process.

[1400] Step 5:

[1401] The server pre-processes the received audio data, including noise filtering and voice normalization, and also pre-processes the collected emotion data to remove outliers.

[1402] Step 6:

[1403] The server runs a speech recognition algorithm to convert the preprocessed audio data into a string of text, from which natural language processing (NLP) is used to extract abusive keywords.

[1404] Step 7:

[1405] The server pre-processes the motion data, filtering the accelerometer data to remove inaccurate data.

[1406] Step 8:

[1407] The server runs a behavioral analysis algorithm to detect patterns of violence based on sudden movements and impacts from the accelerometer data.

[1408] Step 9:

[1409] The server uses an emotion engine to analyze the user's emotional state from the pre-processed emotion data, including analyzing voice tone, speech rate, volume, facial expressions, and body movements.

[1410] Step 10:

[1411] The server integrates information extracted from the voice data, movement data, and emotion data to calculate a stress score for each caregiver, which is quantified based on a set of criteria.

[1412] Step 11:

[1413] The server aggregates the calculated stress scores on a weekly, monthly, and yearly basis, and automatically adjusts the compensation of each care worker based on this.

[1414] Step 12:

[1415] The server connects the results of the compensation adjustment to the payroll system and reflects them in the salaries of each care worker.

[1416] Step 13:

[1417] The server links the quantified stress score to a job information system and makes it publicly available for job seekers to view.

[1418] Step 14:

[1419] Users can browse job information pages and check the stress scores and working environments of each workplace to choose a workplace that suits them.

[1420] As a concrete example, let's say that care worker Yamada wears a wearable device and receives two verbal abuses from a caregiver on Monday ("idiot" and "useless"), and receives one slap on the shoulder. If facial expression recognition detects that Yamada is stressed, the server analyzes this data and reflects a quantified stress score in the payroll system and job information system. This process allows Yamada to receive fair compensation, and job seekers can obtain accurate information about the work environment.

[1421] Example 2

[1422] 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."

[1423] There is a need for a method to visualize and accurately evaluate the stress caused by verbal abuse and violence experienced by caregivers in the nursing care field. Furthermore, there is a need to improve the working environment and reduce staff turnover by appropriately adjusting the compensation of caregivers based on this stress assessment. Furthermore, there is a need for a method to provide information that allows job seekers to understand the working environment in advance.

[1424] 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.

[1425] In this invention, the server includes: means for collecting voice data and movement data received by caregivers wearing wearable devices; means for analyzing the collected voice data to determine whether verbal abuse has occurred; means for analyzing the collected movement data to determine whether violence has occurred; means for collecting emotional data such as facial expressions and body movements to determine the emotional state of the caregiver; means for analyzing the collected emotional data to determine the emotional state of the caregiver; means for quantifying the voice data, movement data, and emotional data based on the analysis results and calculating the caregiver's stress score; means for adjusting the caregiver's compensation based on the quantified information; and means for linking the quantified information to a job information system. This allows for accurate visualization of caregiver stress and adjustment of compensation based on the assessment, thereby improving the working environment and reducing staff turnover. It also allows job seekers to understand the working environment in advance.

[1426] A "wearable device" is a portable device designed to be worn by caregivers and equipped with sensors, microphones, cameras, etc. to collect voice, movement, and emotional data.

[1427] "Audio data" refers to electronically recorded audio information such as conversations that occur during the work of caregivers and environmental sounds.

[1428] "Motion data" refers to electronically recorded data relating to the physical movements and behavior of caregivers, and is collected using an acceleration sensor or the like.

[1429] "Emotional data" refers to electronically recorded data that indicates the emotional state of a caregiver, as inferred from facial expressions, physical movements, heart rate, and electrodermal activity.

[1430] A "voice recognition algorithm" is a technology that analyzes collected voice data, converts it into text data, and detects specific keywords and abusive language.

[1431] "Behavioral analysis algorithms" are technologies that analyze collected behavioral data to detect violent acts and abnormal behavior.

[1432] The "emotion engine" is a technology that analyzes collected emotional data and determines the emotional state of caregivers.

[1433] The "stress score" is a numerical representation of the level of stress of a caregiver calculated based on the results of analyzing voice data, movement data, and emotion data.

[1434] "Remuneration adjustment means" refers to technology and systems for automatically calculating and changing the remuneration of care workers based on stress scores.

[1435] A "job information system" is a system that provides information about working conditions that is made publicly available for job seekers to view.

[1436] This invention aims to visualize stress in nursing care settings and reflect it in compensation, and by combining it with an emotion engine that recognizes the user's emotions, it achieves more accurate stress assessment. This system runs through the steps of collecting data from nursing care staff wearing wearable devices, analyzing it, quantifying it, adjusting compensation, providing information, and recognizing emotions.

[1437] Hardware and Software Configuration

[1438] The terminal uses a wearable device (e.g., a smartwatch or fitness tracker) to collect voice, movement, and emotion data of caregivers in real time. Voice data is collected using the device's built-in microphone, movement data is measured using an acceleration sensor, and emotion data is collected using facial recognition sensors and heart rate sensors.

[1439] The server receives the collected data and performs the following analysis:

[1440] 1. Analysis of audio data

[1441] The server converts the audio data into text using a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API), then detects specific keywords and abusive language.

[1442] 2. Analysis of behavioral data

[1443] The server analyzes the motion data using a behavioral analysis algorithm to detect violent acts and abnormal behavior.

[1444] 3. Emotion Data Analysis

[1445] The server analyzes the emotion data using an emotion engine (e.g., Google Cloud Natural Language API) and determines the emotional state of the caregiver.

[1446] Specific examples

[1447] Let's say that caregiver Yamada wears a wearable device and encounters the following situation.

[1448] Monday

[1449] The device collects data on two occasions when Yamada was subjected to abusive language such as "idiot" and "useless" from users while working.

[1450] The device uses an accelerometer to record the tapping of Yamada's shoulder once.

[1451] The device's facial expression recognition function detects when Yamada is expressing stress.

[1452] The server analyzes this data and determines that there are two cases of verbal abuse, one case of violence, and a high level of stress, and quantifies this data.

[1453] Tuesday

[1454] The device collects data that it received one verbal abuse such as "Stop it!", but records that no violence occurred.

[1455] The device analyzes the voice tone and detects that Yamada is in a state of serious stress.

[1456] From this data, the server determines that one incident of abusive language is a high level of stress and converts it into a numerical value.

[1457] Examples of prompt statements

[1458] An example of a prompt to be input to the generative AI model is as follows:

[1459] Please explain the system that calculates the stress score of caregivers based on the voice data, movement data, and emotion data collected from the caregivers' wearable devices, and automatically adjusts compensation based on that score. The voice data is used to check the presence and frequency of abusive language, and the movement data is used to identify violent acts. The emotion data is used to analyze the stress level using an emotion engine.

[1460] This system will accurately visualize the stress levels of care workers and adjust compensation based on the results, which will improve the working environment and reduce staff turnover. It will also enable job seekers to understand the working environment in advance.

[1461] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1462] Step 1: Data collection

[1463] The terminal collects voice data, movement data, and emotion data of the caregiver wearing the wearable device.

[1464] Input: Voice, movement, and facial expression generated by caregivers

[1465] Specific behavior:

[1466] The device uses a built-in microphone to collect audio data, such as the content of conversations and background sounds.

[1467] The device uses an integrated accelerometer to collect motion data, such as the number of steps taken by the caregiver, posture, and sudden movements.

[1468] The device collects emotional data using built-in facial recognition and heart rate sensors, for example, recording the caregiver's facial expressions and heart rate.

[1469] Output: Collected voice data, movement data, and emotion data

[1470] Step 2: Data Transfer

[1471] The terminal periodically transfers the collected data to the server.

[1472] Input: Collected voice data, movement data, emotion data

[1473] Specific behavior:

[1474] The terminal aggregates the collected data at regular intervals, encrypts it, and sends it to the server.

[1475] Output: Voice data, movement data, and emotion data transferred to the server

[1476] Step 3: Analyzing the audio data

[1477] The server analyzes the received voice data and determines whether or not abusive language is used.

[1478] Input: Audio data sent from the device

[1479] Specific behavior:

[1480] The server converts the voice data into text using a speech recognition algorithm (e.g., Google Cloud Speech-to-Text API).

[1481] The server detects specific keywords and abusive language in the text data.

[1482] Output: Whether or not abusive language was detected and its frequency

[1483] Step 4: Analyzing the behavioral data

[1484] The server analyzes the received motion data and determines whether or not violence has occurred.

[1485] Input: Operation data sent from the device

[1486] Specific behavior:

[1487] The server uses a behavioral analysis algorithm to detect violent acts and abnormal behavior from the motion data.

[1488] For example, violent acts can be identified from abnormal acceleration values ​​or specific movement patterns.

[1489] Output: Whether or not violent acts were detected and their frequency

[1490] Step 5: Analyze the sentiment data

[1491] The server analyzes the received emotion data and determines the emotional state of the caregiver.

[1492] Input: Emotion data sent from the device

[1493] Specific behavior:

[1494] The server uses an emotion engine (e.g., Google Cloud Natural Language API) to classify the emotional state based on data from facial expression recognition sensors and heart rate sensors.

[1495] For example, it can recognize emotions such as smiles, anxiety, and anger from facial expressions, and estimate the level of stress from fluctuations in heart rate.

[1496] Output: Determined emotional state

[1497] Step 6: Calculate your stress score

[1498] The server calculates the stress score of the caregiver based on the analysis results.

[1499] Input: Analyzed voice data, movement data, emotion data

[1500] Specific behavior:

[1501] The server combines the information obtained from each data point and calculates a stress score, taking into account, for example, the number of abusive verbal and violent acts and changes in emotional state.

[1502] Output: Calculated stress score

[1503] Step 7: Compensation Adjustment

[1504] The server adjusts the caregiver's compensation based on the calculated stress score.

[1505] Input: Calculated stress score

[1506] Specific behavior:

[1507] The server calculates the amount of bonus or allowance based on the stress score.

[1508] The calculation results are linked to the payroll system and reflected in monthly pay slips.

[1509] Output: Adjusted reward information

[1510] Step 8: Provide information

[1511] The server links the stress score to a job information system and makes it publicly available for job seekers to view.

[1512] Input: Calculated stress score

[1513] Specific behavior:

[1514] The server uploads the stress score to the job information system and makes it public.

[1515] This allows job seekers to understand the working environment in advance.

[1516] Output: Published stress scores and work environment information

[1517] (Application example 2)

[1518] 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."

[1519] Conventional systems lacked a method for accurately assessing the stress levels of workers in nursing care and factory settings and directly reflecting the results in their compensation. In particular, stress assessments that took emotional data into account were not performed, creating challenges in improving the working environment and increasing worker motivation. Furthermore, there were also insufficient methods for visualizing workers' stress levels in detail and providing information that would allow job seekers to know the working environment in advance.

[1520] 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.

[1521] In this invention, the server includes means for collecting voice data and motion data received by workers wearing wearable devices, means for analyzing the collected voice data and determining whether abusive language has been used, means for analyzing the collected motion data and determining whether violence has been used, means for analyzing the emotion data using an emotion engine and identifying the level of stress, means for quantifying the level of abusive language, violence, and emotion data based on the analysis results, means for adjusting worker compensation based on the quantified information, and means for linking the quantified information to a job information system. This makes it possible to accurately assess worker stress and adjust compensation based on the assessment, as well as enable job seekers to understand the working environment in advance.

[1522] A "wearable device" is a device that can be worn by workers and collects voice data, movement data, emotional data, etc.

[1523] "Audio data" is digital data that records the words spoken by workers and the sounds of their surroundings.

[1524] "Movement data" is digital data collected by sensors on specific physical movements of workers.

[1525] "Abusive language" refers to words that have an insulting or negative connotation towards others.

[1526] "Violence" refers to the act of physically aggressing or inflicting harm on another person.

[1527] An "emotion engine" is software or algorithms that analyze the emotional state of workers based on collected data.

[1528] "Stress level" is a numerical representation of a worker's mental and emotional burden.

[1529] "Quantification" is the process of converting qualitative data into quantitative data that can be treated as such.

[1530] "Remuneration" refers to the monetary compensation received by workers, such as salary, bonuses, and allowances.

[1531] A "job information system" is an online platform where companies and facilities provide job information to job seekers.

[1532] The system for realizing this invention is composed of a wearable device, a server, and a job information system. The processing content of each step, the hardware and software used, and specific examples are explained in detail below.

[1533] Hardware and Software

[1534] Hardware

[1535] Wearable devices: Worn by workers, these devices include microphones to collect voice data, accelerometers to collect movement data, and cameras and sensors to recognize facial expressions.

[1536] Server: Used to analyze collected data and adjust rewards and share information.

[1537] Job Information System: An online platform that provides job seekers with a viewable stress score.

[1538] software

[1539] Speech recognition algorithm: Analyzes collected voice data to determine whether abusive language is present.

[1540] Motion analysis algorithm: Analyzes collected motion data to determine whether violence is present or not.

[1541] Emotion engine: Software for analyzing emotional data and identifying stress levels.

[1542] Compensation Adjustment System: Automatically adjusts workers' compensation based on quantified information.

[1543] Information linkage system: Software for linking quantified information with a job information system.

[1544] Processing steps and functions

[1545] The server processes data, adjusts rewards, and shares information based on the following steps:

[1546] 1. Data collection: Workers wear wearable devices to collect voice, movement, and emotion data, such as microphones, accelerometers, and facial recognition sensors.

[1547] 2. Data Transfer: Collected data is transferred to the server in real time or periodically.

[1548] 3. Data analysis: The server uses a voice recognition algorithm to analyze the voice data and determine whether or not there is abusive language. It also uses a motion analysis algorithm to identify whether or not there is violence. An emotion engine analyzes the level of stress from the emotional data.

[1549] 4. Quantification: Calculate the worker's stress score based on the analyzed voice, movement, and emotional data. For example, a comprehensive score is generated that combines the frequency of verbal and physical abuse with emotional data.

[1550] 5. Compensation Adjustment: Based on the score, the compensation adjustment system will automatically adjust the worker's compensation, which is calculated according to certain criteria.

[1551] 6. Information sharing: The quantified information is transferred to the employment information system via the information sharing system, where it can be viewed by job seekers.

[1552] Specific examples

[1553] For example, suppose a factory worker wears a wearable device while working and collects the following data:

[1554] Voice data: phrases like "I'm not feeling too great today" and stressed voice tones are collected.

[1555] Movement data: Occasionally, certain movement patterns are recorded that indicate strain or fatigue.

[1556] Emotional data: Cameras and sensors detect workers' haggard expressions.

[1557] These data are input into the generative AI model as example prompt sentences as follows:

[1558] Prompt Sentence Examples

[1559] Imagine that Sato is wearing a wearable device while working, and generate the following data.

[1560] Audio data: "I'm not feeling too good today," "I'll try a little harder."

[1561] Movement data: movement patterns that indicate strain or fatigue

[1562] Emotion data: Haggard expression

[1563] These processes enable accurate assessment of worker stress, appropriate compensation adjustments, and visualization of the working environment.

[1564] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1565] Step 1:

[1566] The terminal collects voice, movement, and emotion data from the worker wearing the wearable device. Specifically, it uses a microphone to capture voice data, an accelerometer to measure movement data, and a facial recognition sensor to collect emotion data. This data is then transferred to a server in real time or periodically.

[1567] Step 2:

[1568] The server analyzes the received voice data. Using a voice recognition algorithm, it analyzes specific keywords and tone of voice in the data to determine whether it contains abusive language. The input is the voice data, and the output is a flag indicating whether abusive language is present. For example, if the word "idiot" is included, it is detected as abusive language.

[1569] Step 3:

[1570] The server analyzes the received motion data. Using a motion analysis algorithm, it analyzes specific motion patterns and determines whether a violent act has occurred. The input is the motion data, and the output is a flag indicating whether a violent act has occurred. For example, if a rapid change in acceleration is recorded, it is detected as a violent act.

[1571] Step 4:

[1572] The server uses an emotion engine to analyze the emotion data and calculate the stress level. The input is emotion data, and the output is a numerical value (stress score) that indicates the emotional state. Specifically, if haggard expressions are frequently detected, the stress score will be high.

[1573] Step 5:

[1574] The server integrates the results of the analysis of the voice data, movement data, and emotion data to calculate an overall stress score. Each piece of data is weighted to calculate the overall score. For example, the overall score is created by averaging the frequency of abusive language, the frequency of violent behavior, and emotion data indicating high stress. The inputs are the results of each analysis, and the output is the overall stress score.

[1575] Step 6:

[1576] The server automatically adjusts the worker's compensation based on the calculated overall stress score. It determines whether to increase or decrease compensation depending on the score and transfers the result to the payroll system. Specifically, if the overall score is high, bonuses and allowances are increased. The input is the overall stress score, and the output is the amount of compensation.

[1577] Step 7:

[1578] The server links the quantified stress score to the job information system. The information is made available to job seekers in a viewable format, allowing them to understand the working environment in advance. Specifically, the server displays the stress score for each workplace, helping job seekers select the appropriate workplace. The input is the overall stress score, and the output is the information displayed on the job information system.

[1579] 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.

[1580] 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.

[1581] 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.

[1582] 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.

[1583] 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.

[1584] 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.

[1585] 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).

[1586] 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.

[1587] 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."

[1588] 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.

[1589] 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).

[1590] 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.

[1591] 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.

[1592] 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.

[1593] 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.

[1594] 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.

[1595] 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.

[1596] 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.

[1597] 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.

[1598] 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.

[1599] 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.

[1600] The following is further disclosed regarding the above embodiment.

[1601] (Claim 1)

[1602] A means for collecting voice data and movement data received by a care worker wearing the wearable device;

[1603] A means for analyzing the collected voice data and determining whether abusive language is used;

[1604] A means for analyzing the collected motion data and determining whether or not violence is present;

[1605] A method for quantifying the degree of abusive language and violence based on the analysis results, and

[1606] A means of adjusting nursing staff compensation based on quantified information;

[1607] A means of linking the quantified information to the job information system,

[1608] A system including:

[1609] (Claim 2)

[1610] 2. The system according to claim 1, wherein the voice data is collected using a microphone and the movement data is collected using an acceleration sensor.

[1611] (Claim 3)

[1612] The system according to claim 1, characterized in that the quantified information is compiled weekly, monthly, and annually, and compensation is adjusted based on the compiled information.

[1613] "Example 1"

[1614] (Claim 1)

[1615] A means for collecting voice data and movement data received by a care worker wearing the wearable device;

[1616] A means for analyzing the collected voice data and determining whether abusive language is used;

[1617] A means for analyzing the collected motion data and determining whether or not violence is present;

[1618] A method for quantifying the degree of abusive language and violence based on the analysis results, and

[1619] A means of automatically adjusting the compensation of care workers based on quantified information;

[1620] A means of linking and providing quantified information to the recruitment system,

[1621] A system including:

[1622] (Claim 2)

[1623] 2. The system of claim 1, wherein the collection of voice data utilizes an acoustic sensor and the collection of movement data utilizes a motion sensor.

[1624] (Claim 3)

[1625] The system according to claim 1, characterized in that the quantified information is compiled weekly, monthly, and yearly, and compensation is automatically adjusted based on the quantified information.

[1626] "Application Example 1"

[1627] (Claim 1)

[1628] means for collecting voice and motion data received by a worker wearing the wearable device;

[1629] A means for analyzing the collected voice data and determining whether or not there are words that cause stress;

[1630] means for analyzing the collected motion data to determine whether a motion event indicative of stress has occurred;

[1631] A means for quantifying the stress level based on the analysis results;

[1632] A means of adjusting employee compensation based on quantified information;

[1633] A means of linking the quantified information to the job information system,

[1634] A means of providing quantified information to workers;

[1635] A system including:

[1636] (Claim 2)

[1637] 10. The system of claim 1, wherein the collection of audio data utilizes a microphone and the collection of motion data utilizes an acceleration sensor and a peak detection algorithm.

[1638] (Claim 3)

[1639] The system according to claim 1, characterized in that the quantified information is compiled weekly, monthly, and yearly, and compensation is adjusted based on the compiled information and provided to employees.

[1640] "Example 2: Combining Emotion Engines"

[1641] (Claim 1)

[1642] A means for collecting voice data and movement data received by a care worker wearing the wearable device;

[1643] A means for analyzing the collected voice data and determining whether abusive language is used;

[1644] A means for analyzing the collected motion data and determining whether or not violence is present;

[1645] means for collecting emotional data such as facial expressions and body movements to determine an emotional state;

[1646] A means for analyzing the collected emotional data and determining the emotional state of the caregiver;

[1647] A means for quantifying the voice data, movement data, and emotion data based on the analysis results and calculating a stress score for the caregiver;

[1648] A means of adjusting nursing staff compensation based on quantified information;

[1649] A means of linking the quantified information to the job information system,

[1650] A system including:

[1651] (Claim 2)

[1652] 2. The system according to claim 1, wherein the collection of voice data is performed using a microphone, the collection of movement data is performed using an acceleration sensor, and the collection of emotion data is performed using a facial expression recognition sensor.

[1653] (Claim 3)

[1654] The system according to claim 1, characterized in that the quantified information is compiled weekly, monthly, and annually, and compensation is adjusted based on the compiled information.

[1655] "Application example 2 when combining emotion engines"

[1656] (Claim 1)

[1657] means for collecting audio and motion data received by a worker wearing the wearable device;

[1658] A means for analyzing the collected voice data and determining whether abusive language is used;

[1659] A means for analyzing the collected motion data and determining whether or not violence is present;

[1660] A means for analyzing emotional data using an emotion engine to identify stress levels;

[1661] A method for quantifying the degree of abusive language, violence, and emotional data based on the analysis results, and

[1662] a means of adjusting workers' compensation based on quantified information;

[1663] A means of linking the quantified information to the job information system,

[1664] A system including:

[1665] (Claim 2)

[1666] 2. The system according to claim 1, wherein the collection of voice data is performed using a microphone, the collection of movement data is performed using an acceleration sensor, and the collection of emotion data is performed using a facial expression recognition sensor.

[1667] (Claim 3)

[1668] The system according to claim 1, characterized in that the quantified information is compiled weekly, monthly, and annually, and compensation is adjusted based on the compiled information. [Explanation of symbols]

[1669] 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. A means for collecting voice data and movement data received by a care worker wearing the wearable device; A means for analyzing the collected voice data and determining whether abusive language is used; A means for analyzing the collected motion data and determining whether or not violence is present; A method for quantifying the degree of abusive language and violence based on the analysis results, and A means of adjusting nursing staff compensation based on quantified information; A means of linking the quantified information to the job information system, A system including:

2. 2. The system according to claim 1, wherein the voice data is collected using a microphone, and the motion data is collected using an acceleration sensor.

3. 2. The system according to claim 1, wherein the quantified information is compiled weekly, monthly, and annually, and compensation is adjusted based on the quantified information.

Images