System

Abstract

A system is provided.SOLUTION: A system comprising: a data collection means for collecting facial expressions and sounds of an infant; a data pre-processing means for pre-processing the collected facial expression and sound data; an analysis means for analyzing the pre-processed data to identify a request of the infant; a notification means for converting an analysis result into natural language and notifying a parent; and a learning means for recording a response result of the infant and re-learning based on the result.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] Infants have underdeveloped language skills and are unable to verbally express their needs or state, so parents and guardians must infer responses from their infants' behavior and cries. This situation can be very stressful for parents, delaying appropriate responses. As a result, it can have a negative impact on the infant's health and development. This invention aims to solve this problem by providing a system that analyzes an infant's facial expressions and cries in real time, verbalizes their needs and state, and helps parents respond appropriately. [Means for solving the problem]

[0005] This invention provides a system including a data collection means for collecting an infant's facial expressions and cries, a data preprocessing means for preprocessing the collected facial and cries data, an analysis means for analyzing the preprocessed data to identify the infant's needs, a notification means for converting the analysis results into natural language and notifying the parent, and a learning means for recording the infant's responses and relearning based on those results. The data collection means includes a camera and a microphone and collects the infant's facial expressions and cries in real time. The analysis means identifies the infant's needs using a generative AI model based on the extracted facial and acoustic features. This allows parents to know the infant's needs in real time and take appropriate measures.

[0006] "Data collection means" refers to devices such as cameras and microphones for collecting the infant's facial expressions and sounds.

[0007] "Data pre-processing means" refers to devices and software that perform processes to prepare the collected facial expression and vocalization data for analysis.

[0008] "Analysis tools" refers to generative AI models and machine learning algorithms that identify the infant's condition and needs based on pre-processed data.

[0009] "Notification means" refers to a device or software that converts the analysis results into natural language and notifies the parent or guardian of the results in real time.

[0010] "Learning means" refers to the process or device that accumulates behavioral feedback based on analysis results in a database and retrains the AI ​​model.

[0011] "Facial expression features" refer to features extracted from an infant's facial expressions (e.g., smiling, crying, frown).

[0012] "Acoustic features" refer to sound patterns and characteristics (e.g., frequency spectrum, sound intensity) extracted from infant cries.

[0013] A "generative AI model" is a model that uses machine learning and deep learning techniques to infer an infant's condition and needs from their facial expressions and cries. [Brief explanation of the drawings]

[0014] [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

[0015] 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.

[0016] First, the terms used in the following description will be explained.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 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.

[0025] 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).

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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."

[0035] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, and a learning means.

[0036] Data collection methods

[0037] The data collection means include a camera and microphone to capture the infant's facial expressions and cries. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0038] Data preprocessing measures

[0039] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is applied to extract the position and facial expression features of the infant's face. Acoustic features (e.g., frequency spectrum, sound intensity) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0040] Analysis means

[0041] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0042] Notification means

[0043] The analysis results are converted into natural language on the server. For example, a specific message such as "The baby is hungry. Please give him milk" is generated. The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound.

[0044] Learning tools

[0045] The user (parent) acts according to the advice and records the results in the app. For example, the device records the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0046] Specific examples

[0047] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0048] This system will make it possible to accurately grasp an infant's needs, and is expected to reduce the burden of childcare on parents.

[0049] The processing flow will be explained below.

[0050] Step 1:

[0051] The user (parent) launches the smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time.

[0052] Step 2:

[0053] The device (smartphone) collects video data taken with the camera and audio data recorded with the microphone. This data is temporarily stored on the device and immediately sent to the server.

[0054] Step 3:

[0055] The server receives the video data sent from the device. The video data is divided into frames, and a face detection algorithm is used to identify the baby's face and extract its facial features (e.g., smiling, crying, frown, etc.).

[0056] Step 4:

[0057] The server simultaneously analyzes the received audio data and extracts acoustic features (e.g., frequency spectrum, volume, patterns, etc.). As a result, feature vectors are obtained from both the video and audio.

[0058] Step 5:

[0059] The server inputs this preprocessed data into a generative AI model, which uses facial and acoustic features to identify the infant's needs (e.g., "I'm hungry" or "I'm sleepy") by referring to past learning data.

[0060] Step 6:

[0061] The server converts the analysis results into natural language. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him milk."

[0062] Step 7:

[0063] The server generates a message and sends it to the device, which receives it and notifies the user's smartphone. The notification can be displayed on the screen, sounded, or vibrated to inform the parent.

[0064] Step 8:

[0065] The user receives a notification and acts on the advice, for example, preparing milk and feeding the baby, and then records the results in the app (e.g., the baby's reaction after feeding).

[0066] Step 9:

[0067] The device sends user feedback data to the server, detailing the infant's condition and response results.

[0068] Step 10:

[0069] The server stores the received feedback data in a database. The stored data is periodically used to retrain the AI ​​model, improving the analytical accuracy of the AI ​​model and enabling more accurate judgments.

[0070] Example 1

[0071] 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."

[0072] It is difficult for parents to quickly and accurately grasp their infant's needs and state, which increases the burden of childcare. Furthermore, there is a lack of means to identify appropriate needs from an infant's facial expressions and cries, which can lead to stress due to delayed or incorrect responses. Therefore, a system is needed that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized information to parents.

[0073] 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.

[0074] In this invention, the server includes a data collection means for collecting the facial expressions and cries of the infant, a data preprocessing means for preprocessing the collected facial expression and cries data and extracting the position of the face and acoustic features, an analysis means for identifying the infant's needs using a generative AI model based on the preprocessed data, a notification means for converting the analysis results into natural language and notifying the parent in real time, and a learning means for recording the infant's response results and relearning the generative AI model based on the results. This makes it possible to appropriately and quickly grasp the infant's needs and condition and reduce the burden of childcare on parents.

[0075] The "data collection means" is a device that uses a camera and a microphone to collect the infant's facial expressions and cries and obtain them as digital data.

[0076] The "data preprocessing means" is a process that preprocesses the collected facial expression and sound data, divides the video data into frames to extract facial position and facial expression features, and extracts acoustic features from the audio data.

[0077] The "analysis means" is an algorithm that uses a generative AI model based on pre-processed data to identify the infant's needs and classify them based on facial and acoustic features.

[0078] The "notification means" is a system that converts the analysis results into natural language and notifies parents in real time, and is a function that sends messages via the smartphone screen display, notification sound, etc.

[0079] The "learning method" is the process of recording the infant's response results and relearning the generative AI model based on those results, thereby improving the accuracy of the analysis.

[0080] A "generative AI model" is an artificial intelligence algorithm used to identify an infant's needs using collected and pre-processed data.

[0081] "Facial expression features" are data that indicate changes in facial position and facial expression extracted from video data.

[0082] "Acoustic features" are data that indicate the frequency spectrum, sound intensity, etc. extracted from the audio data.

[0083] "Natural language" refers to the language that humans use on a daily basis, and is the language used to communicate the analysis results to parents in a form that is easy to understand.

[0084] "Relearning" is the process of updating the artificial intelligence algorithm based on collected feedback data to improve analysis accuracy.

[0085] "Real-time" means that the process from data collection to analysis and notification is carried out immediately without delay.

[0086] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, and a learning means.

[0087] Data collection methods

[0088] Users use smartphones or baby monitors to collect real-time video and audio of their infants. This involves using a camera and microphone to capture the infant's movements and cries as digital data. This allows the collection of various facial expressions and cries of the infant, which can then be used for subsequent processing.

[0089] Data preprocessing measures

[0090] The collected digital data is sent from the device (smartphone) to a server. The server first divides the video data into frames and uses a face detection algorithm to extract the position of the infant's face and facial expression features. For example, a library such as OpenCV is used to perform real-time face detection and extract facial expression data. Meanwhile, an FFT (fast Fourier transform) is used on the audio data to extract acoustic features such as frequency spectrum and sound intensity. These preprocessing steps shape the input data for the generative AI model.

[0091] Analysis means

[0092] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. This analysis is performed using machine learning frameworks such as PyTorch. The AI ​​model uses facial and acoustic features as input to classify the infant's needs, such as "hungry" or "sleepy." The accuracy of the analysis can be improved by utilizing past training data.

[0093] Notification means

[0094] The analysis results are converted into natural language on the server. For example, a message such as "The baby is hungry. Please give him milk" is generated. This generated message is sent from the server to the device (smartphone) and notified to the user in real time via the smartphone's screen display and notification sound.

[0095] Learning tools

[0096] The user (parent) acts according to the advice and records the results in the app. For example, they record the baby's reaction after feeding. This data is sent to the server via the device, and the server stores it in a database. The stored data is used to retrain the generative AI model, improving the model's analytical accuracy.

[0097] Specific examples

[0098] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0099] Prompt Sentence Examples

[0100] "When your baby starts crying, open the smartphone app and use the camera and microphone to collect facial expressions and sounds. Press the request button, and the AI ​​model will provide appropriate advice based on the analysis results. After the analysis, you can record and send the behavioral results to the app, which will help improve the accuracy of the AI."

[0101] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0102] Step 1: Data collection

[0103] Server initialization of data collection methods

[0104] Input: The user launches the smartphone app and enables the camera and microphone.

[0105] Specific operation: The user launches the smartphone app and presses the start recording button.

[0106] Output: Real-time video and audio data is collected on a smartphone.

[0107] Step 2: Send data

[0108] Sending data to the server by the device

[0109] Input: Collected video and audio data.

[0110] Specific operation: The terminal divides the collected data into packets and sends them to a server via the Internet.

[0111] Output: Digital data (video and audio data) that arrives at the server.

[0112] Step 3: Data Preprocessing (video)

[0113] Preprocessing of video data by the server

[0114] Input: Video data arriving at the server.

[0115] Specific operation: The server divides the video data into frames and extracts the position and facial expression features of the infant's face using a face detection algorithm (e.g., OpenCV).

[0116] Output: Face position and facial expression feature data for each frame.

[0117] Step 4: Data Preprocessing (Audio)

[0118] Preprocessing of audio data by the server

[0119] Input: The audio data arriving at the server.

[0120] Specific operation: The server applies FFT (Fast Fourier Transform) to the audio data to extract the frequency spectrum and sound intensity.

[0121] Output: Acoustic features of the audio data (frequency spectrum, sound intensity).

[0122] Step 5: Data analysis

[0123] Analysis using AI models generated by the server

[0124] Input: Extracted facial and acoustic feature data.

[0125] How it works: The server inputs the preprocessed data into a generative AI model (e.g., built using PyTorch) to analyze the infant's needs (e.g., "I'm hungry," "I'm sleepy," etc.).

[0126] Output: A request classification result, such as "I'm hungry."

[0127] Step 6: Generate a notification message

[0128] Natural language generation of analysis results by the server

[0129] Input: Requirements classification results from the generative AI model.

[0130] Specific operation: The server uses NLP (natural language processing) technology to convert the request classification results into a message that is easy for the parent to understand (e.g., "The baby is hungry. Please give him milk.").

[0131] Output: A naturalized message.

[0132] Step 7: Notification

[0133] Server sends message to device

[0134] Input: A naturalized message.

[0135] Specific operation: The server uses push notification to send the generated message to the terminal.

[0136] Output: The message arrives at the terminal and the user is notified by the smartphone screen display and notification sound.

[0137] Step 8: User response and feedback

[0138] User response and input of results

[0139] Input: The user acts on the notification (e.g., gives milk) and records the result in the app (e.g., "reaction after giving milk").

[0140] Specific operation: The user enters the corresponding results into the application form and presses the submit button.

[0141] Output: Feedback data recorded on the device.

[0142] Step 9: Sending feedback to the server

[0143] Sending feedback data from the device to the server

[0144] Input: Feedback data recorded on the device.

[0145] Specific operation: The terminal divides the feedback data into packets and sends them to the server via the Internet.

[0146] Output: The feedback data that arrives at the server.

[0147] Step 10: Retrain

[0148] Retraining of the generated AI model by the server

[0149] Input: Collected feedback data.

[0150] Specific operation: The server periodically ingests feedback data in batches and retrains the generative AI model.

[0151] Output: Improved analysis accuracy with updated AI model.

[0152] (Application example 1)

[0153] 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."

[0154] When driving a car with an infant, parents are required to respond quickly to their infant's requests while concentrating on driving. However, it is difficult to check the infant's facial expressions or cries while driving, which can result in delayed responses. Furthermore, if parents are unable to understand and respond appropriately to their infant's requests while driving, the infant's safety and comfort may be compromised. A system is needed to improve this situation and respond quickly and accurately to infant requests even while driving.

[0155] 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.

[0156] In this invention, the server includes a data collection means, a data preprocessing means, an analysis means, a notification means, a learning means, a control means for controlling the driving of the vehicle based on the notification results from the analysis means, and a display means for displaying the content of a request to the driver when a request from the infant occurs while driving. This allows parents to quickly understand the infant's request and respond appropriately even while driving.

[0157] The "data collection means" is a device that includes a camera and a microphone to collect the infant's facial expressions and sounds.

[0158] The "data pre-processing means" is a processing device for converting the collected facial expression and vocalization data into a format suitable for analysis.

[0159] The "analysis means" is a function for identifying infants' needs based on pre-processed data, and utilizes a generative AI model.

[0160] The "notification means" is a function for converting the analysis results into natural language and notifying the parents.

[0161] The "learning means" is a function that records the infant's response results and re-learns the system based on those results.

[0162] The "control means" is a device for controlling the driving of the vehicle based on the results of notification by the analysis means.

[0163] The "display means" is a device for displaying the request content to the driver when a request is made by the baby while driving.

[0164] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, a learning means, a control means, and a display means.

[0165] Data collection methods

[0166] The data collection means includes a camera and microphone to capture the facial expressions and cries of the infant. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0167] Data preprocessing measures

[0168] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is applied to extract the position and facial expression features of the infant's face. Acoustic features (e.g., frequency spectrum, volume) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0169] Analysis means

[0170] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0171] Notification means

[0172] The analysis results are converted into natural language on the server. For example, a specific message such as "The baby is hungry. Please give him milk" is generated. The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound.

[0173] Learning tools

[0174] The user (parent) acts according to the advice and records the results in the app. For example, recording the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0175] Control means

[0176] The server controls the driving of the car based on the results of the analysis. For example, if a baby is crying, the server can automatically slow down the car or instruct it to stop in a safe place.

[0177] Display means

[0178] If a request is made by an infant while driving, the server displays the request to the driver, using the vehicle's dashboard monitor or head-up display so that the driver can check it without looking away.

[0179] Specific examples

[0180] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0181] An example of a prompt statement is, "What should you do if your baby is crying?"

[0182] This system allows parents to quickly and accurately grasp their infant's needs and respond appropriately even while driving.

[0183] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0184] Step 1:

[0185] The user (parent) launches the smartphone app and uses the camera and microphone to collect the baby's facial expressions and cries.

[0186] Input: Video and audio data captured from the smartphone camera and microphone.

[0187] Output: Video and audio data of infants collected in real time.

[0188] Step 2:

[0189] The terminal transmits the collected video and audio data to a server.

[0190] Input: Real-time collected video and audio data of infants.

[0191] Output: Video and audio data of the infant sent to the server.

[0192] Step 3:

[0193] The server divides the received video data into frames and applies a face detection algorithm to extract the position and facial expression features of the infant's face.

[0194] Input: Video data sent to the server.

[0195] Output: Data containing face position and expression features.

[0196] Step 4:

[0197] The server extracts acoustic features (e.g., frequency spectrum, volume) from the received audio data.

[0198] Input: The audio data sent to the server.

[0199] Output: Data containing acoustic features.

[0200] Step 5:

[0201] The server analyzes the infant's needs using a generative AI model based on the extracted facial and acoustic features.

[0202] Input: Data including facial position, facial expression features, and acoustic features.

[0203] Output: Data identifying the infant's needs (e.g., hungry, sleepy, etc.).

[0204] Step 6:

[0205] The server generates a natural language message based on the infant's request and transmits it to the terminal.

[0206] Input: Data identifying the infant's needs.

[0207] Output: A natural language message such as "The baby is hungry. Please give him milk."

[0208] Step 7:

[0209] The device will display the received message on the parent's smartphone and notify the parent with a notification sound or vibration.

[0210] Input: A natural language message sent by the server.

[0211] Output: Messages and notifications that appear on your smartphone screen.

[0212] Step 8:

[0213] Parents follow the app's instructions to provide the necessary care for their infant, and then record the results (e.g., the infant's reaction after feeding) in the app.

[0214] Input: Data about parental behavior and its outcomes.

[0215] Output: Feedback data including the infant's reaction after feeding.

[0216] Step 9:

[0217] The terminal transmits the recorded feedback data to the server.

[0218] Input: Recorded feedback data.

[0219] Output: Feedback data sent to the server.

[0220] Step 10:

[0221] The server stores the feedback data in a database and uses it to retrain the AI ​​model.

[0222] Input: Feedback data.

[0223] Output: The updated AI model.

[0224] Step 11:

[0225] The server controls the driving of the car based on the results of the analysis. For example, if a baby is crying, the server can automatically slow down the car or instruct it to stop in a safe place.

[0226] Input: Notification results from analysis means.

[0227] Output: Car driving control instructions.

[0228] Step 12:

[0229] When an infant makes a request while driving, the server displays the request to the driver. This display is made on the vehicle's dashboard monitor or head-up display so that the driver can check it without looking away.

[0230] Input: Data about the infant's needs.

[0231] Output: Requests displayed on dashboard monitor or head-up display.

[0232] 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.

[0233] This invention is a system that analyzes an infant's facial expressions and cries to verbalize its requests and state, and further recognizes the emotional state of the user (parent) and adjusts the content of notifications. This system includes data collection means, data preprocessing means, analysis means, notification means, learning means, and an emotion engine.

[0234] Data collection methods

[0235] The data collection means include a camera and microphone to capture the infant's facial expressions and cries. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0236] Data preprocessing measures

[0237] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is used to extract the position and facial expression of the infant's face. Acoustic features (e.g., frequency spectrum, volume) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0238] Analysis means

[0239] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0240] Notification means

[0241] The analysis results are converted into natural language on the server. For example, if the AI ​​model determines that the baby is hungry, it generates a specific message such as, "The baby is hungry. Please give him some milk." The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound. Furthermore, the content of the notification can be adjusted taking into account the user's emotional state.

[0242] Emotion Engine

[0243] An emotion engine is a device or software that analyzes a user's facial expressions and behavior to recognize their emotions. It identifies the user's stress level and emotional state (e.g., joy, anger, sadness). The emotion engine identifies the user's emotional state, and the notification mechanism adjusts the message to take this into account. For example, if the user is stressed, the notification content will include additional psychological support.

[0244] Learning tools

[0245] The user (parent) acts according to the advice and records the results in the app. For example, the device records the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0246] Specific examples

[0247] The user (parent) launches a smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time, and this data is sent to a server. The server extracts facial features from the video data and acoustic features from the audio data. A generative AI model analyzes these features and identifies the baby's request as "hungry."

[0248] As a result, the server generates a message saying, "The baby is hungry. Please give him milk," and sends it to the device. The device then notifies the smartphone, and the user receives this notification and prepares the milk. The user then records the result in the app, and the device sends the feedback data to the server. The server uses this data to retrain the AI ​​model.

[0249] Furthermore, if the emotion engine analyzes the user's emotional state and determines that the user is feeling stressed, it will send a message containing psychological support such as, "The baby is hungry. Please feed him / her. Please stay calm and don't panic."

[0250] This system makes it possible to accurately grasp the needs of infants, reducing the burden of childcare on parents and providing appropriate support according to the user's emotional state.

[0251] The processing flow will be explained below.

[0252] Step 1:

[0253] The user (parent) launches the smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time.

[0254] Step 2:

[0255] The device (smartphone) collects video data taken with its camera and audio data recorded with its microphone. The collected data is temporarily stored on the device and then sent to a server.

[0256] Step 3:

[0257] The server receives the video data sent from the device, divides the video data into frames, and uses a face detection algorithm to identify the baby's face and extract its facial features (e.g., smiling, crying, frown, etc.).

[0258] Step 4:

[0259] The server simultaneously analyzes the received audio data and extracts acoustic features (e.g., frequency spectrum, volume, patterns, etc.). As a result, feature vectors are obtained from both the video and audio.

[0260] Step 5:

[0261] The server inputs the preprocessed data into the generative AI model, which uses facial and acoustic features to refer to past learning data to identify the infant's needs (e.g., "I'm hungry" or "I'm sleepy") and obtains the analysis results.

[0262] Step 6:

[0263] The server converts the analysis results into natural language. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him milk."

[0264] Step 7:

[0265] The server sends the generated message to the device, which receives it and notifies the user's smartphone.

[0266] Step 8:

[0267] To enable the device to recognize the user's emotional state, the device uses the smartphone's front camera to collect facial expression data, which is then sent to a server.

[0268] Step 9:

[0269] The server uses an emotion engine to analyze the user's facial expression data, which determines the user's stress level and emotional state (e.g., happy, anger, sadness).

[0270] Step 10:

[0271] The server adjusts the notification content based on the user's emotional state. For example, if the user is under stress, the server generates a message containing psychological support such as, "The baby is hungry. Please feed him / her. Please stay calm and don't panic."

[0272] Step 11:

[0273] The server sends the adjusted notification message to the device, which then notifies the user via a screen display, notification sound, or vibration.

[0274] Step 12:

[0275] The user receives a notification and acts on the advice, for example, preparing milk and feeding the baby, and then records the results in the app (for example, the baby's reaction after feeding).

[0276] Step 13:

[0277] The terminal transmits feedback data from the user to the server, including the user's behavior and the infant's reactions.

[0278] Step 14:

[0279] The server stores the received feedback data in a database, which is then periodically used to retrain the AI ​​model, improving the analysis accuracy of the AI ​​model and the overall performance of the system.

[0280] Example 2

[0281] 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."

[0282] Accurately understanding an infant's needs is a significant burden for many parents and often causes parenting stress. Furthermore, depending on the parent's emotional state, they may not be able to respond appropriately. To solve these problems, there is a need for a system that analyzes an infant's facial expressions and cries, verbalizes their needs, and recognizes the parent's emotional state to adjust the content of notifications accordingly.

[0283] 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.

[0284] In this invention, the server includes a data collection means for collecting the facial expressions and cries of the infant, a data preprocessing means for preprocessing the collected facial expression and cries data, an analysis means for analyzing the preprocessed data to identify the infant's needs, a notification means for converting the analysis results into natural language and notifying the parent, a learning means for recording the infant's responses and relearning based on the results, and an emotion engine for recognizing the user's emotional state and adjusting the content of the notification. This makes it possible to accurately grasp the infant's needs and to provide appropriate notifications according to the parent's emotional state.

[0285] "Infant facial expressions and vocalizations" refers to the patterns of an infant's facial movements, facial expressions, and oral sounds.

[0286] "Data collection means" refers to devices or systems for collecting the infant's facial expressions and cries. Specifically, cameras and microphones are used.

[0287] "Data pre-processing methods" refer to techniques and processes used to convert collected raw data into a form suitable for analysis, such as video frame segmentation or acoustic feature extraction.

[0288] "Analytical tools" refer to algorithms and techniques used to identify infant needs based on pre-processed data, specifically generative AI models.

[0289] "Notification means" refers to the technology or system that converts the analysis results into natural language and notifies the parents, for example by displaying a message on a smartphone.

[0290] "Learning methods" refer to the techniques and methods used to record the infant's responses and re-learn based on those results. Specifically, the feedback data is used to improve the accuracy of the AI ​​model.

[0291] An "emotion engine" is a technology or system that recognizes a user's emotional state and adjusts notification content accordingly, for example by analyzing the user's facial expressions and voice.

[0292] MODE FOR CARRYING OUT THE INVENTION

[0293] This invention is a system that analyzes an infant's facial expressions and cries to verbalize its requests and state, and further recognizes the emotional state of the user (parent) and adjusts the content of notifications. This system includes data collection means, data preprocessing means, analysis means, notification means, learning means, and an emotion engine.

[0294] Data collection methods

[0295] Users use smartphones or baby monitors to collect video and audio of their infants in real time. This is achieved by using the smartphone's camera and microphone as data collection means. Through this operation, the infant's movements and cries are stored as digital data on the device.

[0296] Data preprocessing measures

[0297] The device sends the collected video and audio data to a server. The server divides the video data into frames and uses a face detection algorithm to extract the position and facial expression features of the infant's face. Acoustic features (such as frequency spectrum and sound intensity) are also extracted from the audio data. Libraries such as OpenCV, Dlib, and Librosa are used for this preprocessing. Preprocessing allows the data to be formatted before being input into the AI ​​model.

[0298] Analysis means

[0299] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if the facial and acoustic features match the pattern of "hungry," the AI ​​model will determine that the infant is "hungry." This analysis is performed using deep learning frameworks such as TensorFlow and PyTorch.

[0300] Notification means

[0301] The analysis results are converted into natural language on the server. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him some milk." The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is displayed on the smartphone screen and uses a notification sound, and is sent via Firebase Cloud Messaging or Apple Push Notification Service.

[0302] Emotion Engine

[0303] An emotion engine is a device or software that analyzes a user's facial expressions and behavior to recognize their emotional state. It uses emotion recognition libraries such as TensorFlow and OpenFace. For example, it identifies the user's stress level and emotional state (e.g., joy, anger, sadness). The emotion engine identifies the user's emotional state, and the notification method adjusts the message to take this into account. For example, if the user is stressed, an additional message such as "Please remain calm and respond calmly" is included.

[0304] Learning tools

[0305] The user acts according to the notification and records the results in the app. For example, the user can record the baby's reaction after feeding the baby. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the generative AI model, improving its analytical accuracy.

[0306] Specific examples

[0307] The user launches a smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time, and these data are sent to a server. The server extracts facial features from the video data and acoustic features from the audio data. A generative AI model analyzes these features and identifies the baby's request as "hungry." As a result, the server generates a message saying, "The baby is hungry. Please feed him / her." and sends it to the device. The device notifies the smartphone, and the user prepares the milk. The user then records the result in the app, and the device sends feedback data to the server. The server uses this data to retrain the generative AI model. Furthermore, an emotion engine analyzes the user's emotional state. If it determines that the user is stressed, it sends a message containing psychological support, such as, "The baby is hungry. Please feed him / her. Please stay calm and respond calmly."

[0308] Prompt Sentence Examples

[0309] "Please explain the data you would like to input into a generative AI model for a system that analyzes a baby's facial expressions and vocalizations to determine its needs and notify the parents."

[0310] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0311] Step 1:

[0312] The user launches the smartphone app and points the camera at the baby. When the user starts recording with the app, the smartphone's camera and microphone begin collecting video and audio of the baby. The input at this stage is the baby's real-time facial expressions and cries, and the output is video and audio data stored on the smartphone as digital data.

[0313] Step 2:

[0314] The smartphone (device) sends the collected video and audio data to a server using a communication method, specifically Wi-Fi or mobile data communication. The input at this stage is the video and audio data stored on the smartphone, and the output is digital data transferred to the server.

[0315] Step 3:

[0316] The server divides the received video data into frames and extracts the infant's facial position and facial expression features using a face detection algorithm (e.g., OpenCV or Dlib). It also extracts acoustic features (e.g., frequency spectrum and sound intensity) from the audio data using an audio processing library such as Librosa. The input at this stage is the video and audio data sent to the server, and the output is the processed facial and acoustic features.

[0317] Step 4:

[0318] The server inputs the preprocessed facial and acoustic features into a generative AI model to analyze the infant's needs. Specifically, the model performs analysis using deep learning frameworks such as TensorFlow and PyTorch to identify the infant's needs (e.g., "I'm hungry"). The input at this stage is the preprocessed feature data, and the output is the analyzed infant's needs.

[0319] Step 5:

[0320] The server converts the analysis results into natural language. This is achieved using an NLP engine such as GPT-3 or BERT. For example, if the analysis result is "hungry," the server generates a message saying "The baby is hungry. Please give him milk." The input at this stage is the analysis result as the infant's request, and the output is a notification message in natural language.

[0321] Step 6:

[0322] The server then sends the generated notification message to the device (smartphone). This is done via Firebase Cloud Messaging (FCM) or Apple Push Notification Service (APNs). The user receives the message in real time on their smartphone and is notified by a notification sound or a display on the screen. The input at this stage is a natural language notification message, and the output is a notification displayed on the smartphone.

[0323] Step 7:

[0324] The server records the user's facial expressions and behavior and analyzes their emotional state using an emotion engine (such as TensorFlow or OpenFace). If the server determines that the user is feeling stressed, it includes an additional psychological support message in the notification. The input of this stage is the user's facial expressions and voice data, and the output is a tailored notification message.

[0325] Step 8:

[0326] The user acts according to the notification and records the results in the app. For example, the user can enter the baby's reaction after feeding into the app. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the generative AI model. The input at this stage is the feedback data recorded by the user in the app, and the output is the retraining data stored on the server.

[0327] (Application example 2)

[0328] 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."

[0329] A system is needed that can not only accurately grasp the needs and status of infants and notify parents in real time, but also recognize the parents' emotional state and provide appropriate notifications according to the situation. In particular, there is a need for a system that can respond quickly to infants' needs while taking into consideration the parents' stress levels and emotional state, thereby reducing the burden of childcare and providing higher quality childcare support.

[0330] 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.

[0331] In this invention, the server includes data collection means for collecting the facial expressions and cries of the infant, data preprocessing means for preprocessing the collected facial expression and cries data, analysis means for analyzing the preprocessed data to identify the infant's needs, notification means for converting the analysis results into natural language and notifying the guardian, learning means for recording the infant's response results and relearning based on those results, and an emotion engine for recognizing the user's emotional state and adjusting the content of the notification. This makes it possible to accurately grasp the infant's needs and provide appropriate support according to the guardian's emotional state.

[0332] The "data collection means" is a device that includes a camera and a microphone to collect the infant's facial expressions and sounds.

[0333] The "data preprocessing means" is a means for executing a process to prepare the collected facial expression and vocalization data in a form suitable for analysis.

[0334] "Analysis means" means a means of using a generative AI model to identify the needs of an infant based on pre-processed data.

[0335] The "notification means" is a means for converting the analysis results into natural language and notifying the parents.

[0336] The "learning means" is a means for recording the infant's response results and using that information to improve the system's analytical accuracy.

[0337] The "emotion engine" is a means of recognizing the parent's emotional state and adjusting the content of notifications based on that information.

[0338] A "generative AI model" is an artificial intelligence model that identifies the needs and conditions of infants and young children based on collected and pre-processed data.

[0339] The system for implementing this invention collects and analyzes the facial expressions and cries of infants, and then notifies parents appropriately, thereby reducing the burden of childcare. It also recognizes the parents' emotional state and adjusts the content of notifications based on that information. This system utilizes various hardware and software and is implemented as follows:

[0340] Hardware and software used

[0341] Hardware: smartphone, smart glasses, server, camera and microphone.

[0342] Software: Data collection application (for smartphone / smart glasses), data pre-processing algorithm (server), emotion recognition model (generative AI model), notification application (smartphone).

[0343] System Overview

[0344] Data collection methods

[0345] The system uses a device (such as a smartphone or smart glasses) to collect the infant's facial expressions and cries. This uses a camera and microphone, and the collected video and audio data is sent to a server via the internet.

[0346] Data preprocessing measures

[0347] The server splits the video data into frames and performs face detection using the OpenCV library. It also extracts acoustic features such as frequency spectrum and volume from the audio data. These preprocessing steps prepare the data in a format suitable for analysis.

[0348] Analysis means

[0349] Based on the pre-processed data, a generative AI model is used to identify the infant's needs, such as whether they are hungry or sleepy. This analysis also utilizes past training data, so accuracy improves over time.

[0350] Notification means

[0351] The analysis results are converted into natural language on the server and sent to the parent's device. Notifications are displayed on the smartphone screen or via voice notification. The emotion engine also recognizes the parent's emotional state and adjusts the content of the notification based on that information. For example, if the parent is feeling stressed, a message containing psychological support will be sent.

[0352] Learning tools

[0353] Parents act on the advice provided and record their results in the app. This feedback data is sent to the server and used to retrain the model, further improving its analysis accuracy.

[0354] Specific example explanation

[0355] 1. Data Collection

[0356] Parents launch the smartphone app and take a picture of their baby's face with the camera. Audio data is also collected, and this data is sent to a server in real time.

[0357] 2. Data Preprocessing

[0358] The server preprocesses the data that arrives, extracting facial features from video data and acoustic features from audio data.

[0359] 3. Analysis

[0360] A generative AI model is used to analyze an infant's needs, for example, to identify the need to "be hungry."

[0361] 4. Notification

[0362] A message is generated and sent to the parent's smartphone saying, "Your baby is hungry. Please feed him / her." If the parent's emotional state indicates stress, an additional message is included saying, "Please stay calm."

[0363] Prompt Sentence Examples

[0364] Prompt: "Create a system that detects delivery workers' fatigue and suggests breaks. Collect facial expression data from the delivery workers using a camera and analyze the fatigue indicators using a model."

[0365] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0366] Step 1: Data collection

[0367] The user (guardian) starts the smartphone app and takes a picture of the baby's face with the camera. At the same time, audio is collected with the microphone. The collected video data (facial expressions) and audio data (cries) are sent from the device to the server in real time. This makes the input data available on the server.

[0368] Step 2: Data Preprocessing

[0369] The server splits the received video data into frames using the OpenCV library and performs face detection, extracting facial features. At the same time, acoustic features such as frequency spectrum and volume are extracted from the audio data. Once preprocessing is complete, these feature data become available as input data for analysis.

[0370] Step 3: Data analysis

[0371] The server inputs the preprocessed facial and acoustic features into a generative AI model. The generative AI model analyzes this data and identifies the infant's needs. For example, the request "I'm hungry" may be identified. The analysis results are output as a natural language message indicating a specific response.

[0372] Step 4: Notification

[0373] The server then sends the generated natural language message to the parent's smartphone. Notifications are displayed on the screen or sounded. The emotion engine then analyzes the parent's emotional state, and if it detects stress, it adds psychological support to the message. For example, it may include a message saying, "Please stay calm."

[0374] Step 5: Gather feedback

[0375] The user (guardian) takes action according to the notification and then records the results in the smartphone app. For example, they input "the baby's reaction after giving milk." This feedback data is sent from the device to the server.

[0376] Step 6: Retrain the model

[0377] The server saves the collected feedback data in a database and uses it to retrain the generative AI model, thereby improving the accuracy of analysis. As the feedback data accumulated sequentially is used as training data for the model, the accuracy of the model improves over time.

[0378] Based on this processing step, the system can accurately grasp the needs of infants and provide appropriate notifications that take into account the emotional state of the caregiver, thereby reducing the burden of childcare on the caregiver.

[0379] 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.

[0380] 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.

[0381] 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.

[0382] [Second embodiment]

[0383] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0384] 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.

[0385] 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).

[0386] 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.

[0387] 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.

[0388] 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).

[0389] 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.

[0390] 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.

[0391] 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.

[0392] 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.

[0393] 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.

[0394] 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."

[0395] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, and a learning means.

[0396] Data collection methods

[0397] The data collection means include a camera and microphone to capture the infant's facial expressions and cries. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0398] Data preprocessing measures

[0399] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is applied to extract the position and facial expression features of the infant's face. Acoustic features (e.g., frequency spectrum, sound intensity) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0400] Analysis means

[0401] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0402] Notification means

[0403] The analysis results are converted into natural language on the server. For example, a specific message such as "The baby is hungry. Please give him milk" is generated. The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound.

[0404] Learning tools

[0405] The user (parent) acts according to the advice and records the results in the app. For example, the device records the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0406] Specific examples

[0407] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0408] This system will make it possible to accurately grasp an infant's needs, and is expected to reduce the burden of childcare on parents.

[0409] The processing flow will be explained below.

[0410] Step 1:

[0411] The user (parent) launches the smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time.

[0412] Step 2:

[0413] The device (smartphone) collects video data taken with the camera and audio data recorded with the microphone. This data is temporarily stored on the device and immediately sent to the server.

[0414] Step 3:

[0415] The server receives the video data sent from the device. The video data is divided into frames, and a face detection algorithm is used to identify the baby's face and extract its facial features (e.g., smiling, crying, frown, etc.).

[0416] Step 4:

[0417] The server simultaneously analyzes the received audio data and extracts acoustic features (e.g., frequency spectrum, volume, patterns, etc.). As a result, feature vectors are obtained from both the video and audio.

[0418] Step 5:

[0419] The server inputs this preprocessed data into a generative AI model, which uses facial and acoustic features to identify the infant's needs (e.g., "I'm hungry" or "I'm sleepy") by referring to past learning data.

[0420] Step 6:

[0421] The server converts the analysis results into natural language. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him milk."

[0422] Step 7:

[0423] The server generates a message and sends it to the device, which receives it and notifies the user's smartphone. The notification can be displayed on the screen, sounded, or vibrated to inform the parent.

[0424] Step 8:

[0425] The user receives a notification and acts on the advice, for example, preparing milk and feeding the baby, and then records the results in the app (e.g., the baby's reaction after feeding).

[0426] Step 9:

[0427] The device sends user feedback data to the server, detailing the infant's condition and response results.

[0428] Step 10:

[0429] The server stores the received feedback data in a database. The stored data is periodically used to retrain the AI ​​model, improving the analytical accuracy of the AI ​​model and enabling more accurate judgments.

[0430] Example 1

[0431] 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."

[0432] It is difficult for parents to quickly and accurately grasp their infant's needs and state, which increases the burden of childcare. Furthermore, there is a lack of means to identify appropriate needs from an infant's facial expressions and cries, which can lead to stress due to delayed or incorrect responses. Therefore, a system is needed that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized information to parents.

[0433] 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.

[0434] In this invention, the server includes a data collection means for collecting the facial expressions and cries of the infant, a data preprocessing means for preprocessing the collected facial expression and cries data and extracting the position of the face and acoustic features, an analysis means for identifying the infant's needs using a generative AI model based on the preprocessed data, a notification means for converting the analysis results into natural language and notifying the parent in real time, and a learning means for recording the infant's response results and relearning the generative AI model based on the results. This makes it possible to appropriately and quickly grasp the infant's needs and condition and reduce the burden of childcare on parents.

[0435] The "data collection means" is a device that uses a camera and a microphone to collect the infant's facial expressions and cries and obtain them as digital data.

[0436] The "data preprocessing means" is a process that preprocesses the collected facial expression and sound data, divides the video data into frames to extract facial position and facial expression features, and extracts acoustic features from the audio data.

[0437] The "analysis means" is an algorithm that uses a generative AI model based on pre-processed data to identify the infant's needs and classify them based on facial and acoustic features.

[0438] The "notification means" is a system that converts the analysis results into natural language and notifies parents in real time, and is a function that sends messages via the smartphone screen display, notification sound, etc.

[0439] The "learning method" is the process of recording the infant's response results and relearning the generative AI model based on those results, thereby improving the accuracy of the analysis.

[0440] A "generative AI model" is an artificial intelligence algorithm used to identify an infant's needs using collected and pre-processed data.

[0441] "Facial expression features" are data that indicate changes in facial position and facial expression extracted from video data.

[0442] "Acoustic features" are data that indicate the frequency spectrum, sound intensity, etc. extracted from the audio data.

[0443] "Natural language" refers to the language that humans use on a daily basis, and is the language used to communicate the analysis results to parents in a form that is easy to understand.

[0444] "Relearning" is the process of updating the artificial intelligence algorithm based on collected feedback data to improve analysis accuracy.

[0445] "Real-time" means that the process from data collection to analysis and notification is carried out immediately without delay.

[0446] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, and a learning means.

[0447] Data collection methods

[0448] Users use smartphones or baby monitors to collect real-time video and audio of their infants. This involves using a camera and microphone to capture the infant's movements and cries as digital data. This allows the collection of various facial expressions and cries of the infant, which can then be used for subsequent processing.

[0449] Data preprocessing measures

[0450] The collected digital data is sent from the device (smartphone) to a server. The server first divides the video data into frames and uses a face detection algorithm to extract the position of the infant's face and facial expression features. For example, a library such as OpenCV is used to perform real-time face detection and extract facial expression data. Meanwhile, an FFT (fast Fourier transform) is used on the audio data to extract acoustic features such as frequency spectrum and sound intensity. These preprocessing steps shape the input data for the generative AI model.

[0451] Analysis means

[0452] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. This analysis is performed using machine learning frameworks such as PyTorch. The AI ​​model uses facial and acoustic features as input to classify the infant's needs, such as "hungry" or "sleepy." The accuracy of the analysis can be improved by utilizing past training data.

[0453] Notification means

[0454] The analysis results are converted into natural language on the server. For example, a message such as "The baby is hungry. Please give him milk" is generated. This generated message is sent from the server to the device (smartphone) and notified to the user in real time via the smartphone's screen display and notification sound.

[0455] Learning tools

[0456] The user (parent) acts according to the advice and records the results in the app. For example, they record the baby's reaction after feeding. This data is sent to the server via the device, and the server stores it in a database. The stored data is used to retrain the generative AI model, improving the model's analytical accuracy.

[0457] Specific examples

[0458] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0459] Prompt Sentence Examples

[0460] "When your baby starts crying, open the smartphone app and use the camera and microphone to collect facial expressions and sounds. Press the request button, and the AI ​​model will provide appropriate advice based on the analysis results. After the analysis, you can record and send the behavioral results to the app, which will help improve the accuracy of the AI."

[0461] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0462] Step 1: Data collection

[0463] Server initialization of data collection methods

[0464] Input: The user launches the smartphone app and enables the camera and microphone.

[0465] Specific operation: The user launches the smartphone app and presses the start recording button.

[0466] Output: Real-time video and audio data is collected on a smartphone.

[0467] Step 2: Send data

[0468] Sending data to the server by the device

[0469] Input: Collected video and audio data.

[0470] Specific operation: The terminal divides the collected data into packets and sends them to a server via the Internet.

[0471] Output: Digital data (video and audio data) that arrives at the server.

[0472] Step 3: Data Preprocessing (video)

[0473] Preprocessing of video data by the server

[0474] Input: Video data arriving at the server.

[0475] Specific operation: The server divides the video data into frames and extracts the position and facial expression features of the infant's face using a face detection algorithm (e.g., OpenCV).

[0476] Output: Face position and facial expression feature data for each frame.

[0477] Step 4: Data Preprocessing (Audio)

[0478] Preprocessing of audio data by the server

[0479] Input: The audio data arriving at the server.

[0480] Specific operation: The server applies FFT (Fast Fourier Transform) to the audio data to extract the frequency spectrum and sound intensity.

[0481] Output: Acoustic features of the audio data (frequency spectrum, sound intensity).

[0482] Step 5: Data analysis

[0483] Analysis using AI models generated by the server

[0484] Input: Extracted facial and acoustic feature data.

[0485] How it works: The server inputs the preprocessed data into a generative AI model (e.g., built using PyTorch) to analyze the infant's needs (e.g., "I'm hungry," "I'm sleepy," etc.).

[0486] Output: A request classification result, such as "I'm hungry."

[0487] Step 6: Generate a notification message

[0488] Natural language generation of analysis results by the server

[0489] Input: Requirements classification results from the generative AI model.

[0490] Specific operation: The server uses NLP (natural language processing) technology to convert the request classification results into a message that is easy for the parent to understand (e.g., "The baby is hungry. Please give him milk.").

[0491] Output: A naturalized message.

[0492] Step 7: Notification

[0493] Server sends message to device

[0494] Input: A naturalized message.

[0495] Specific operation: The server uses push notification to send the generated message to the terminal.

[0496] Output: The message arrives at the terminal and the user is notified by the smartphone screen display and notification sound.

[0497] Step 8: User response and feedback

[0498] User response and input of results

[0499] Input: The user acts on the notification (e.g., gives milk) and records the result in the app (e.g., "reaction after giving milk").

[0500] Specific operation: The user enters the corresponding results into the application form and presses the submit button.

[0501] Output: Feedback data recorded on the device.

[0502] Step 9: Sending feedback to the server

[0503] Sending feedback data from the device to the server

[0504] Input: Feedback data recorded on the device.

[0505] Specific operation: The terminal divides the feedback data into packets and sends them to the server via the Internet.

[0506] Output: The feedback data that arrives at the server.

[0507] Step 10: Retrain

[0508] Retraining of the generated AI model by the server

[0509] Input: Collected feedback data.

[0510] Specific operation: The server periodically ingests feedback data in batches and retrains the generative AI model.

[0511] Output: Improved analysis accuracy with updated AI model.

[0512] (Application example 1)

[0513] 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."

[0514] When driving a car with an infant, parents are required to respond quickly to their infant's requests while concentrating on driving. However, it is difficult to check the infant's facial expressions or cries while driving, which can result in delayed responses. Furthermore, if parents are unable to understand and respond appropriately to their infant's requests while driving, the infant's safety and comfort may be compromised. A system is needed to improve this situation and respond quickly and accurately to infant requests even while driving.

[0515] 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.

[0516] In this invention, the server includes a data collection means, a data preprocessing means, an analysis means, a notification means, a learning means, a control means for controlling the driving of the vehicle based on the notification results from the analysis means, and a display means for displaying the content of a request to the driver when a request from the infant occurs while driving. This allows parents to quickly understand the infant's request and respond appropriately even while driving.

[0517] The "data collection means" is a device that includes a camera and a microphone to collect the infant's facial expressions and sounds.

[0518] The "data pre-processing means" is a processing device for converting the collected facial expression and vocalization data into a format suitable for analysis.

[0519] The "analysis means" is a function for identifying infants' needs based on pre-processed data, and utilizes a generative AI model.

[0520] The "notification means" is a function for converting the analysis results into natural language and notifying the parents.

[0521] The "learning means" is a function that records the infant's response results and re-learns the system based on those results.

[0522] The "control means" is a device for controlling the driving of the vehicle based on the results of notification by the analysis means.

[0523] The "display means" is a device for displaying the request content to the driver when a request is made by the baby while driving.

[0524] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, a learning means, a control means, and a display means.

[0525] Data collection methods

[0526] The data collection means includes a camera and microphone to capture the facial expressions and cries of the infant. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0527] Data preprocessing measures

[0528] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is applied to extract the position and facial expression features of the infant's face. Acoustic features (e.g., frequency spectrum, volume) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0529] Analysis means

[0530] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0531] Notification means

[0532] The analysis results are converted into natural language on the server. For example, a specific message such as "The baby is hungry. Please give him milk" is generated. The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound.

[0533] Learning tools

[0534] The user (parent) acts according to the advice and records the results in the app. For example, recording the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0535] Control means

[0536] The server controls the driving of the car based on the results of the analysis. For example, if a baby is crying, the server can automatically slow down the car or instruct it to stop in a safe place.

[0537] Display means

[0538] If a request is made by an infant while driving, the server displays the request to the driver, using the vehicle's dashboard monitor or head-up display so that the driver can check it without looking away.

[0539] Specific examples

[0540] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0541] An example of a prompt statement is, "What should you do if your baby is crying?"

[0542] This system allows parents to quickly and accurately grasp their infant's needs and respond appropriately even while driving.

[0543] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0544] Step 1:

[0545] The user (parent) launches the smartphone app and uses the camera and microphone to collect the baby's facial expressions and cries.

[0546] Input: Video and audio data captured from the smartphone camera and microphone.

[0547] Output: Video and audio data of infants collected in real time.

[0548] Step 2:

[0549] The terminal transmits the collected video and audio data to a server.

[0550] Input: Real-time collected video and audio data of infants.

[0551] Output: Video and audio data of the infant sent to the server.

[0552] Step 3:

[0553] The server divides the received video data into frames and applies a face detection algorithm to extract the position and facial expression features of the infant's face.

[0554] Input: Video data sent to the server.

[0555] Output: Data containing face position and expression features.

[0556] Step 4:

[0557] The server extracts acoustic features (e.g., frequency spectrum, volume) from the received audio data.

[0558] Input: The audio data sent to the server.

[0559] Output: Data containing acoustic features.

[0560] Step 5:

[0561] The server analyzes the infant's needs using a generative AI model based on the extracted facial and acoustic features.

[0562] Input: Data including facial position, facial expression features, and acoustic features.

[0563] Output: Data identifying the infant's needs (e.g., hungry, sleepy, etc.).

[0564] Step 6:

[0565] The server generates a natural language message based on the infant's request and transmits it to the terminal.

[0566] Input: Data identifying the infant's needs.

[0567] Output: A natural language message such as "The baby is hungry. Please give him milk."

[0568] Step 7:

[0569] The device will display the received message on the parent's smartphone and notify the parent with a notification sound or vibration.

[0570] Input: A natural language message sent by the server.

[0571] Output: Messages and notifications that appear on your smartphone screen.

[0572] Step 8:

[0573] Parents follow the app's instructions to provide the necessary care for their infant, and then record the results (e.g., the infant's reaction after feeding) in the app.

[0574] Input: Data about parental behavior and its outcomes.

[0575] Output: Feedback data including the infant's reaction after feeding.

[0576] Step 9:

[0577] The terminal transmits the recorded feedback data to the server.

[0578] Input: Recorded feedback data.

[0579] Output: Feedback data sent to the server.

[0580] Step 10:

[0581] The server stores the feedback data in a database and uses it to retrain the AI ​​model.

[0582] Input: Feedback data.

[0583] Output: The updated AI model.

[0584] Step 11:

[0585] The server controls the driving of the car based on the results of the analysis. For example, if a baby is crying, the server can automatically slow down the car or instruct it to stop in a safe place.

[0586] Input: Notification results from analysis means.

[0587] Output: Car driving control instructions.

[0588] Step 12:

[0589] When an infant makes a request while driving, the server displays the request to the driver. This display is made on the vehicle's dashboard monitor or head-up display so that the driver can check it without looking away.

[0590] Input: Data about the infant's needs.

[0591] Output: Requests displayed on dashboard monitor or head-up display.

[0592] 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.

[0593] This invention is a system that analyzes an infant's facial expressions and cries to verbalize its requests and state, and further recognizes the emotional state of the user (parent) and adjusts the content of notifications. This system includes data collection means, data preprocessing means, analysis means, notification means, learning means, and an emotion engine.

[0594] Data collection methods

[0595] The data collection means include a camera and microphone to capture the infant's facial expressions and cries. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0596] Data preprocessing measures

[0597] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is used to extract the position and facial expression of the infant's face. Acoustic features (e.g., frequency spectrum, volume) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0598] Analysis means

[0599] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0600] Notification means

[0601] The analysis results are converted into natural language on the server. For example, if the AI ​​model determines that the baby is hungry, it generates a specific message such as, "The baby is hungry. Please give him some milk." The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound. Furthermore, the content of the notification can be adjusted taking into account the user's emotional state.

[0602] Emotion Engine

[0603] An emotion engine is a device or software that analyzes a user's facial expressions and behavior to recognize their emotions. It identifies the user's stress level and emotional state (e.g., joy, anger, sadness). The emotion engine identifies the user's emotional state, and the notification mechanism adjusts the message to take this into account. For example, if the user is stressed, the notification content will include additional psychological support.

[0604] Learning tools

[0605] The user (parent) acts according to the advice and records the results in the app. For example, the device records the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0606] Specific examples

[0607] The user (parent) launches a smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time, and this data is sent to a server. The server extracts facial features from the video data and acoustic features from the audio data. A generative AI model analyzes these features and identifies the baby's request as "hungry."

[0608] As a result, the server generates a message saying, "The baby is hungry. Please give him milk," and sends it to the device. The device then notifies the smartphone, and the user receives this notification and prepares the milk. The user then records the result in the app, and the device sends the feedback data to the server. The server uses this data to retrain the AI ​​model.

[0609] Furthermore, if the emotion engine analyzes the user's emotional state and determines that the user is feeling stressed, it will send a message containing psychological support such as, "The baby is hungry. Please feed him / her. Please stay calm and don't panic."

[0610] This system makes it possible to accurately grasp the needs of infants, reducing the burden of childcare on parents and providing appropriate support according to the user's emotional state.

[0611] The processing flow will be explained below.

[0612] Step 1:

[0613] The user (parent) launches the smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time.

[0614] Step 2:

[0615] The device (smartphone) collects video data taken with its camera and audio data recorded with its microphone. The collected data is temporarily stored on the device and then sent to a server.

[0616] Step 3:

[0617] The server receives the video data sent from the device, divides the video data into frames, and uses a face detection algorithm to identify the baby's face and extract its facial features (e.g., smiling, crying, frown, etc.).

[0618] Step 4:

[0619] The server simultaneously analyzes the received audio data and extracts acoustic features (e.g., frequency spectrum, volume, patterns, etc.). As a result, feature vectors are obtained from both the video and audio.

[0620] Step 5:

[0621] The server inputs the preprocessed data into the generative AI model, which uses facial and acoustic features to refer to past learning data to identify the infant's needs (e.g., "I'm hungry" or "I'm sleepy") and obtains the analysis results.

[0622] Step 6:

[0623] The server converts the analysis results into natural language. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him milk."

[0624] Step 7:

[0625] The server sends the generated message to the device, which receives it and notifies the user's smartphone.

[0626] Step 8:

[0627] To enable the device to recognize the user's emotional state, the device uses the smartphone's front camera to collect facial expression data, which is then sent to a server.

[0628] Step 9:

[0629] The server uses an emotion engine to analyze the user's facial expression data, which determines the user's stress level and emotional state (e.g., happy, anger, sadness).

[0630] Step 10:

[0631] The server adjusts the notification content based on the user's emotional state. For example, if the user is under stress, the server generates a message containing psychological support such as, "The baby is hungry. Please feed him / her. Please stay calm and don't panic."

[0632] Step 11:

[0633] The server sends the adjusted notification message to the device, which then notifies the user via a screen display, notification sound, or vibration.

[0634] Step 12:

[0635] The user receives a notification and acts on the advice, for example, preparing milk and feeding the baby, and then records the results in the app (for example, the baby's reaction after feeding).

[0636] Step 13:

[0637] The terminal transmits feedback data from the user to the server, including the user's behavior and the infant's reactions.

[0638] Step 14:

[0639] The server stores the received feedback data in a database, which is then periodically used to retrain the AI ​​model, improving the analysis accuracy of the AI ​​model and the overall performance of the system.

[0640] Example 2

[0641] 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."

[0642] Accurately understanding an infant's needs is a significant burden for many parents and often causes parenting stress. Furthermore, depending on the parent's emotional state, they may not be able to respond appropriately. To solve these problems, there is a need for a system that analyzes an infant's facial expressions and cries, verbalizes their needs, and recognizes the parent's emotional state to adjust the content of notifications accordingly.

[0643] 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.

[0644] In this invention, the server includes a data collection means for collecting the facial expressions and cries of the infant, a data preprocessing means for preprocessing the collected facial expression and cries data, an analysis means for analyzing the preprocessed data to identify the infant's needs, a notification means for converting the analysis results into natural language and notifying the parent, a learning means for recording the infant's responses and relearning based on the results, and an emotion engine for recognizing the user's emotional state and adjusting the content of the notification. This makes it possible to accurately grasp the infant's needs and to provide appropriate notifications according to the parent's emotional state.

[0645] "Infant facial expressions and vocalizations" refers to the patterns of an infant's facial movements, facial expressions, and oral sounds.

[0646] "Data collection means" refers to devices or systems for collecting the infant's facial expressions and cries. Specifically, cameras and microphones are used.

[0647] "Data pre-processing methods" refer to techniques and processes used to convert collected raw data into a form suitable for analysis, such as video frame segmentation or acoustic feature extraction.

[0648] "Analytical tools" refer to algorithms and techniques used to identify infant needs based on pre-processed data, specifically generative AI models.

[0649] "Notification means" refers to the technology or system that converts the analysis results into natural language and notifies the parents, for example by displaying a message on a smartphone.

[0650] "Learning methods" refer to the techniques and methods used to record the infant's responses and re-learn based on those results. Specifically, the feedback data is used to improve the accuracy of the AI ​​model.

[0651] An "emotion engine" is a technology or system that recognizes a user's emotional state and adjusts notification content accordingly, for example by analyzing the user's facial expressions and voice.

[0652] MODE FOR CARRYING OUT THE INVENTION

[0653] This invention is a system that analyzes an infant's facial expressions and cries to verbalize its requests and state, and further recognizes the emotional state of the user (parent) and adjusts the content of notifications. This system includes data collection means, data preprocessing means, analysis means, notification means, learning means, and an emotion engine.

[0654] Data collection methods

[0655] Users use smartphones or baby monitors to collect video and audio of their infants in real time. This is achieved by using the smartphone's camera and microphone as data collection means. Through this operation, the infant's movements and cries are stored as digital data on the device.

[0656] Data preprocessing measures

[0657] The device sends the collected video and audio data to a server. The server divides the video data into frames and uses a face detection algorithm to extract the position and facial expression features of the infant's face. Acoustic features (such as frequency spectrum and sound intensity) are also extracted from the audio data. Libraries such as OpenCV, Dlib, and Librosa are used for this preprocessing. Preprocessing allows the data to be formatted before being input into the AI ​​model.

[0658] Analysis means

[0659] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if the facial and acoustic features match the pattern of "hungry," the AI ​​model will determine that the infant is "hungry." This analysis is performed using deep learning frameworks such as TensorFlow and PyTorch.

[0660] Notification means

[0661] The analysis results are converted into natural language on the server. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him some milk." The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is displayed on the smartphone screen and uses a notification sound, and is sent via Firebase Cloud Messaging or Apple Push Notification Service.

[0662] Emotion Engine

[0663] An emotion engine is a device or software that analyzes a user's facial expressions and behavior to recognize their emotional state. It uses emotion recognition libraries such as TensorFlow and OpenFace. For example, it identifies the user's stress level and emotional state (e.g., joy, anger, sadness). The emotion engine identifies the user's emotional state, and the notification method adjusts the message to take this into account. For example, if the user is stressed, an additional message such as "Please remain calm and respond calmly" is included.

[0664] Learning tools

[0665] The user acts according to the notification and records the results in the app. For example, the user can record the baby's reaction after feeding the baby. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the generative AI model, improving its analytical accuracy.

[0666] Specific examples

[0667] The user launches a smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time, and these data are sent to a server. The server extracts facial features from the video data and acoustic features from the audio data. A generative AI model analyzes these features and identifies the baby's request as "hungry." As a result, the server generates a message saying, "The baby is hungry. Please feed him / her." and sends it to the device. The device notifies the smartphone, and the user prepares the milk. The user then records the result in the app, and the device sends feedback data to the server. The server uses this data to retrain the generative AI model. Furthermore, an emotion engine analyzes the user's emotional state. If it determines that the user is stressed, it sends a message containing psychological support, such as, "The baby is hungry. Please feed him / her. Please stay calm and respond calmly."

[0668] Prompt Sentence Examples

[0669] "Please explain the data you would like to input into a generative AI model for a system that analyzes a baby's facial expressions and vocalizations to determine its needs and notify the parents."

[0670] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0671] Step 1:

[0672] The user launches the smartphone app and points the camera at the baby. When the user starts recording with the app, the smartphone's camera and microphone begin collecting video and audio of the baby. The input at this stage is the baby's real-time facial expressions and cries, and the output is video and audio data stored on the smartphone as digital data.

[0673] Step 2:

[0674] The smartphone (device) sends the collected video and audio data to a server using a communication method, specifically Wi-Fi or mobile data communication. The input at this stage is the video and audio data stored on the smartphone, and the output is digital data transferred to the server.

[0675] Step 3:

[0676] The server divides the received video data into frames and extracts the infant's facial position and facial expression features using a face detection algorithm (e.g., OpenCV or Dlib). It also extracts acoustic features (e.g., frequency spectrum and sound intensity) from the audio data using an audio processing library such as Librosa. The input at this stage is the video and audio data sent to the server, and the output is the processed facial and acoustic features.

[0677] Step 4:

[0678] The server inputs the preprocessed facial and acoustic features into a generative AI model to analyze the infant's needs. Specifically, the model performs analysis using deep learning frameworks such as TensorFlow and PyTorch to identify the infant's needs (e.g., "I'm hungry"). The input at this stage is the preprocessed feature data, and the output is the analyzed infant's needs.

[0679] Step 5:

[0680] The server converts the analysis results into natural language. This is achieved using an NLP engine such as GPT-3 or BERT. For example, if the analysis result is "hungry," the server generates a message saying "The baby is hungry. Please give him milk." The input at this stage is the analysis result as the infant's request, and the output is a notification message in natural language.

[0681] Step 6:

[0682] The server then sends the generated notification message to the device (smartphone). This is done via Firebase Cloud Messaging (FCM) or Apple Push Notification Service (APNs). The user receives the message in real time on their smartphone and is notified by a notification sound or a display on the screen. The input at this stage is a natural language notification message, and the output is a notification displayed on the smartphone.

[0683] Step 7:

[0684] The server records the user's facial expressions and behavior and analyzes their emotional state using an emotion engine (such as TensorFlow or OpenFace). If the server determines that the user is feeling stressed, it includes an additional psychological support message in the notification. The input of this stage is the user's facial expressions and voice data, and the output is a tailored notification message.

[0685] Step 8:

[0686] The user acts according to the notification and records the results in the app. For example, the user can enter the baby's reaction after feeding into the app. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the generative AI model. The input at this stage is the feedback data recorded by the user in the app, and the output is the retraining data stored on the server.

[0687] (Application example 2)

[0688] 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."

[0689] A system is needed that can not only accurately grasp the needs and status of infants and notify parents in real time, but also recognize the parents' emotional state and provide appropriate notifications according to the situation. In particular, there is a need for a system that can respond quickly to infants' needs while taking into consideration the parents' stress levels and emotional state, thereby reducing the burden of childcare and providing higher quality childcare support.

[0690] 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.

[0691] In this invention, the server includes data collection means for collecting the facial expressions and cries of the infant, data preprocessing means for preprocessing the collected facial expression and cries data, analysis means for analyzing the preprocessed data to identify the infant's needs, notification means for converting the analysis results into natural language and notifying the guardian, learning means for recording the infant's response results and relearning based on those results, and an emotion engine for recognizing the user's emotional state and adjusting the content of the notification. This makes it possible to accurately grasp the infant's needs and provide appropriate support according to the guardian's emotional state.

[0692] The "data collection means" is a device that includes a camera and a microphone to collect the infant's facial expressions and sounds.

[0693] The "data preprocessing means" is a means for executing a process to prepare the collected facial expression and vocalization data in a form suitable for analysis.

[0694] "Analysis means" means a means of using a generative AI model to identify the needs of an infant based on pre-processed data.

[0695] The "notification means" is a means for converting the analysis results into natural language and notifying the parents.

[0696] The "learning means" is a means for recording the infant's response results and using that information to improve the system's analytical accuracy.

[0697] The "emotion engine" is a means of recognizing the parent's emotional state and adjusting the content of notifications based on that information.

[0698] A "generative AI model" is an artificial intelligence model that identifies the needs and conditions of infants and young children based on collected and pre-processed data.

[0699] The system for implementing this invention collects and analyzes the facial expressions and cries of infants, and then notifies parents appropriately, thereby reducing the burden of childcare. It also recognizes the parents' emotional state and adjusts the content of notifications based on that information. This system utilizes various hardware and software and is implemented as follows:

[0700] Hardware and software used

[0701] Hardware: smartphone, smart glasses, server, camera and microphone.

[0702] Software: Data collection application (for smartphone / smart glasses), data pre-processing algorithm (server), emotion recognition model (generative AI model), notification application (smartphone).

[0703] System Overview

[0704] Data collection methods

[0705] The system uses a device (such as a smartphone or smart glasses) to collect the infant's facial expressions and cries. This uses a camera and microphone, and the collected video and audio data is sent to a server via the internet.

[0706] Data preprocessing measures

[0707] The server splits the video data into frames and performs face detection using the OpenCV library. It also extracts acoustic features such as frequency spectrum and volume from the audio data. These preprocessing steps prepare the data in a format suitable for analysis.

[0708] Analysis means

[0709] Based on the pre-processed data, a generative AI model is used to identify the infant's needs, such as whether they are hungry or sleepy. This analysis also utilizes past training data, so accuracy improves over time.

[0710] Notification means

[0711] The analysis results are converted into natural language on the server and sent to the parent's device. Notifications are displayed on the smartphone screen or via voice notification. The emotion engine also recognizes the parent's emotional state and adjusts the content of the notification based on that information. For example, if the parent is feeling stressed, a message containing psychological support will be sent.

[0712] Learning tools

[0713] Parents act on the advice provided and record their results in the app. This feedback data is sent to the server and used to retrain the model, further improving its analysis accuracy.

[0714] Specific example explanation

[0715] 1. Data Collection

[0716] Parents launch the smartphone app and take a picture of their baby's face with the camera. Audio data is also collected, and this data is sent to a server in real time.

[0717] 2. Data Preprocessing

[0718] The server preprocesses the data that arrives, extracting facial features from video data and acoustic features from audio data.

[0719] 3. Analysis

[0720] A generative AI model is used to analyze an infant's needs, for example, to identify the need to "be hungry."

[0721] 4. Notification

[0722] A message is generated and sent to the parent's smartphone saying, "Your baby is hungry. Please feed him / her." If the parent's emotional state indicates stress, an additional message is included saying, "Please stay calm."

[0723] Prompt Sentence Examples

[0724] Prompt: "Create a system that detects delivery workers' fatigue and suggests breaks. Collect facial expression data from the delivery workers using a camera and analyze the fatigue indicators using a model."

[0725] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0726] Step 1: Data collection

[0727] The user (guardian) starts the smartphone app and takes a picture of the baby's face with the camera. At the same time, audio is collected with the microphone. The collected video data (facial expressions) and audio data (cries) are sent from the device to the server in real time. This makes the input data available on the server.

[0728] Step 2: Data Preprocessing

[0729] The server splits the received video data into frames using the OpenCV library and performs face detection, extracting facial features. At the same time, acoustic features such as frequency spectrum and volume are extracted from the audio data. Once preprocessing is complete, these feature data become available as input data for analysis.

[0730] Step 3: Data analysis

[0731] The server inputs the preprocessed facial and acoustic features into a generative AI model. The generative AI model analyzes this data and identifies the infant's needs. For example, the request "I'm hungry" may be identified. The analysis results are output as a natural language message indicating a specific response.

[0732] Step 4: Notification

[0733] The server then sends the generated natural language message to the parent's smartphone. Notifications are displayed on the screen or sounded. The emotion engine then analyzes the parent's emotional state, and if it detects stress, it adds psychological support to the message. For example, it may include a message saying, "Please stay calm."

[0734] Step 5: Gather feedback

[0735] The user (guardian) takes action according to the notification and then records the results in the smartphone app. For example, they input "the baby's reaction after giving milk." This feedback data is sent from the device to the server.

[0736] Step 6: Retrain the model

[0737] The server saves the collected feedback data in a database and uses it to retrain the generative AI model, thereby improving the accuracy of analysis. As the feedback data accumulated sequentially is used as training data for the model, the accuracy of the model improves over time.

[0738] Based on this processing step, the system can accurately grasp the needs of infants and provide appropriate notifications that take into account the emotional state of the caregiver, thereby reducing the burden of childcare on the caregiver.

[0739] 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.

[0740] 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.

[0741] 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.

[0742] [Third embodiment]

[0743] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0744] 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.

[0745] 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).

[0746] 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.

[0747] 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.

[0748] 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).

[0749] 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.

[0750] 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.

[0751] 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.

[0752] 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.

[0753] 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.

[0754] 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."

[0755] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, and a learning means.

[0756] Data collection methods

[0757] The data collection means include a camera and microphone to capture the infant's facial expressions and cries. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0758] Data preprocessing measures

[0759] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is applied to extract the position and facial expression features of the infant's face. Acoustic features (e.g., frequency spectrum, sound intensity) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0760] Analysis means

[0761] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0762] Notification means

[0763] The analysis results are converted into natural language on the server. For example, a specific message such as "The baby is hungry. Please give him milk" is generated. The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound.

[0764] Learning tools

[0765] The user (parent) acts according to the advice and records the results in the app. For example, the device records the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0766] Specific examples

[0767] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0768] This system will make it possible to accurately grasp an infant's needs, and is expected to reduce the burden of childcare on parents.

[0769] The processing flow will be explained below.

[0770] Step 1:

[0771] The user (parent) launches the smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time.

[0772] Step 2:

[0773] The device (smartphone) collects video data taken with the camera and audio data recorded with the microphone. This data is temporarily stored on the device and immediately sent to the server.

[0774] Step 3:

[0775] The server receives the video data sent from the device. The video data is divided into frames, and a face detection algorithm is used to identify the baby's face and extract its facial features (e.g., smiling, crying, frown, etc.).

[0776] Step 4:

[0777] The server simultaneously analyzes the received audio data and extracts acoustic features (e.g., frequency spectrum, volume, patterns, etc.). As a result, feature vectors are obtained from both the video and audio.

[0778] Step 5:

[0779] The server inputs this preprocessed data into a generative AI model, which uses facial and acoustic features to identify the infant's needs (e.g., "I'm hungry" or "I'm sleepy") by referring to past learning data.

[0780] Step 6:

[0781] The server converts the analysis results into natural language. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him milk."

[0782] Step 7:

[0783] The server generates a message and sends it to the device, which receives it and notifies the user's smartphone. The notification can be displayed on the screen, sounded, or vibrated to inform the parent.

[0784] Step 8:

[0785] The user receives a notification and acts on the advice, for example, preparing milk and feeding the baby, and then records the results in the app (e.g., the baby's reaction after feeding).

[0786] Step 9:

[0787] The device sends user feedback data to the server, detailing the infant's condition and response results.

[0788] Step 10:

[0789] The server stores the received feedback data in a database. The stored data is periodically used to retrain the AI ​​model, improving the analytical accuracy of the AI ​​model and enabling more accurate judgments.

[0790] Example 1

[0791] 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."

[0792] It is difficult for parents to quickly and accurately grasp their infant's needs and state, which increases the burden of childcare. Furthermore, there is a lack of means to identify appropriate needs from an infant's facial expressions and cries, which can lead to stress due to delayed or incorrect responses. Therefore, a system is needed that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized information to parents.

[0793] 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.

[0794] In this invention, the server includes a data collection means for collecting the facial expressions and cries of the infant, a data preprocessing means for preprocessing the collected facial expression and cries data and extracting the position of the face and acoustic features, an analysis means for identifying the infant's needs using a generative AI model based on the preprocessed data, a notification means for converting the analysis results into natural language and notifying the parent in real time, and a learning means for recording the infant's response results and relearning the generative AI model based on the results. This makes it possible to appropriately and quickly grasp the infant's needs and condition and reduce the burden of childcare on parents.

[0795] The "data collection means" is a device that uses a camera and a microphone to collect the infant's facial expressions and cries and obtain them as digital data.

[0796] The "data preprocessing means" is a process that preprocesses the collected facial expression and sound data, divides the video data into frames to extract facial position and facial expression features, and extracts acoustic features from the audio data.

[0797] The "analysis means" is an algorithm that uses a generative AI model based on pre-processed data to identify the infant's needs and classify them based on facial and acoustic features.

[0798] The "notification means" is a system that converts the analysis results into natural language and notifies parents in real time, and is a function that sends messages via the smartphone screen display, notification sound, etc.

[0799] The "learning method" is the process of recording the infant's response results and relearning the generative AI model based on those results, thereby improving the accuracy of the analysis.

[0800] A "generative AI model" is an artificial intelligence algorithm used to identify an infant's needs using collected and pre-processed data.

[0801] "Facial expression features" are data that indicate changes in facial position and facial expression extracted from video data.

[0802] "Acoustic features" are data that indicate the frequency spectrum, sound intensity, etc. extracted from the audio data.

[0803] "Natural language" refers to the language that humans use on a daily basis, and is the language used to communicate the analysis results to parents in a form that is easy to understand.

[0804] "Relearning" is the process of updating the artificial intelligence algorithm based on collected feedback data to improve analysis accuracy.

[0805] "Real-time" means that the process from data collection to analysis and notification is carried out immediately without delay.

[0806] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, and a learning means.

[0807] Data collection methods

[0808] Users use smartphones or baby monitors to collect real-time video and audio of their infants. This involves using a camera and microphone to capture the infant's movements and cries as digital data. This allows the collection of various facial expressions and cries of the infant, which can then be used for subsequent processing.

[0809] Data preprocessing measures

[0810] The collected digital data is sent from the device (smartphone) to a server. The server first divides the video data into frames and uses a face detection algorithm to extract the position of the infant's face and facial expression features. For example, a library such as OpenCV is used to perform real-time face detection and extract facial expression data. Meanwhile, an FFT (fast Fourier transform) is used on the audio data to extract acoustic features such as frequency spectrum and sound intensity. These preprocessing steps shape the input data for the generative AI model.

[0811] Analysis means

[0812] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. This analysis is performed using machine learning frameworks such as PyTorch. The AI ​​model uses facial and acoustic features as input to classify the infant's needs, such as "hungry" or "sleepy." The accuracy of the analysis can be improved by utilizing past training data.

[0813] Notification means

[0814] The analysis results are converted into natural language on the server. For example, a message such as "The baby is hungry. Please give him milk" is generated. This generated message is sent from the server to the device (smartphone) and notified to the user in real time via the smartphone's screen display and notification sound.

[0815] Learning tools

[0816] The user (parent) acts according to the advice and records the results in the app. For example, they record the baby's reaction after feeding. This data is sent to the server via the device, and the server stores it in a database. The stored data is used to retrain the generative AI model, improving the model's analytical accuracy.

[0817] Specific examples

[0818] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0819] Prompt Sentence Examples

[0820] "When your baby starts crying, open the smartphone app and use the camera and microphone to collect facial expressions and sounds. Press the request button, and the AI ​​model will provide appropriate advice based on the analysis results. After the analysis, you can record and send the behavioral results to the app, which will help improve the accuracy of the AI."

[0821] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0822] Step 1: Data collection

[0823] Server initialization of data collection methods

[0824] Input: The user launches the smartphone app and enables the camera and microphone.

[0825] Specific operation: The user launches the smartphone app and presses the start recording button.

[0826] Output: Real-time video and audio data is collected on a smartphone.

[0827] Step 2: Send data

[0828] Sending data to the server by the device

[0829] Input: Collected video and audio data.

[0830] Specific operation: The terminal divides the collected data into packets and sends them to a server via the Internet.

[0831] Output: Digital data (video and audio data) that arrives at the server.

[0832] Step 3: Data Preprocessing (video)

[0833] Preprocessing of video data by the server

[0834] Input: Video data arriving at the server.

[0835] Specific operation: The server divides the video data into frames and extracts the position and facial expression features of the infant's face using a face detection algorithm (e.g., OpenCV).

[0836] Output: Face position and facial expression feature data for each frame.

[0837] Step 4: Data Preprocessing (Audio)

[0838] Preprocessing of audio data by the server

[0839] Input: The audio data arriving at the server.

[0840] Specific operation: The server applies FFT (Fast Fourier Transform) to the audio data to extract the frequency spectrum and sound intensity.

[0841] Output: Acoustic features of the audio data (frequency spectrum, sound intensity).

[0842] Step 5: Data analysis

[0843] Analysis using AI models generated by the server

[0844] Input: Extracted facial and acoustic feature data.

[0845] How it works: The server inputs the preprocessed data into a generative AI model (e.g., built using PyTorch) to analyze the infant's needs (e.g., "I'm hungry," "I'm sleepy," etc.).

[0846] Output: A request classification result, such as "I'm hungry."

[0847] Step 6: Generate a notification message

[0848] Natural language generation of analysis results by the server

[0849] Input: Requirements classification results from the generative AI model.

[0850] Specific operation: The server uses NLP (natural language processing) technology to convert the request classification results into a message that is easy for the parent to understand (e.g., "The baby is hungry. Please give him milk.").

[0851] Output: A naturalized message.

[0852] Step 7: Notification

[0853] Server sends message to device

[0854] Input: A naturalized message.

[0855] Specific operation: The server uses push notification to send the generated message to the terminal.

[0856] Output: The message arrives at the terminal and the user is notified by the smartphone screen display and notification sound.

[0857] Step 8: User response and feedback

[0858] User response and input of results

[0859] Input: The user acts on the notification (e.g., gives milk) and records the result in the app (e.g., "reaction after giving milk").

[0860] Specific operation: The user enters the corresponding results into the application form and presses the submit button.

[0861] Output: Feedback data recorded on the device.

[0862] Step 9: Sending feedback to the server

[0863] Sending feedback data from the device to the server

[0864] Input: Feedback data recorded on the device.

[0865] Specific operation: The terminal divides the feedback data into packets and sends them to the server via the Internet.

[0866] Output: The feedback data that arrives at the server.

[0867] Step 10: Retrain

[0868] Retraining of the generated AI model by the server

[0869] Input: Collected feedback data.

[0870] Specific operation: The server periodically ingests feedback data in batches and retrains the generative AI model.

[0871] Output: Improved analysis accuracy with updated AI model.

[0872] (Application example 1)

[0873] 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."

[0874] When driving a car with an infant, parents are required to respond quickly to their infant's requests while concentrating on driving. However, it is difficult to check the infant's facial expressions or cries while driving, which can result in delayed responses. Furthermore, if parents are unable to understand and respond appropriately to their infant's requests while driving, the infant's safety and comfort may be compromised. A system is needed to improve this situation and respond quickly and accurately to infant requests even while driving.

[0875] 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.

[0876] In this invention, the server includes a data collection means, a data preprocessing means, an analysis means, a notification means, a learning means, a control means for controlling the driving of the vehicle based on the notification results from the analysis means, and a display means for displaying the content of a request to the driver when a request from the infant occurs while driving. This allows parents to quickly understand the infant's request and respond appropriately even while driving.

[0877] The "data collection means" is a device that includes a camera and a microphone to collect the infant's facial expressions and sounds.

[0878] The "data pre-processing means" is a processing device for converting the collected facial expression and vocalization data into a format suitable for analysis.

[0879] The "analysis means" is a function for identifying infants' needs based on pre-processed data, and utilizes a generative AI model.

[0880] The "notification means" is a function for converting the analysis results into natural language and notifying the parents.

[0881] The "learning means" is a function that records the infant's response results and re-learns the system based on those results.

[0882] The "control means" is a device for controlling the driving of the vehicle based on the results of notification by the analysis means.

[0883] The "display means" is a device for displaying the request content to the driver when a request is made by the baby while driving.

[0884] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, a learning means, a control means, and a display means.

[0885] Data collection methods

[0886] The data collection means includes a camera and microphone to capture the facial expressions and cries of the infant. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0887] Data preprocessing measures

[0888] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is applied to extract the position and facial expression features of the infant's face. Acoustic features (e.g., frequency spectrum, volume) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0889] Analysis means

[0890] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0891] Notification means

[0892] The analysis results are converted into natural language on the server. For example, a specific message such as "The baby is hungry. Please give him milk" is generated. The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound.

[0893] Learning tools

[0894] The user (parent) acts according to the advice and records the results in the app. For example, recording the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0895] Control means

[0896] The server controls the driving of the car based on the results of the analysis. For example, if a baby is crying, the server can automatically slow down the car or instruct it to stop in a safe place.

[0897] Display means

[0898] If a request is made by an infant while driving, the server displays the request to the driver, using the vehicle's dashboard monitor or head-up display so that the driver can check it without looking away.

[0899] Specific examples

[0900] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[0901] An example of a prompt statement is, "What should you do if your baby is crying?"

[0902] This system allows parents to quickly and accurately grasp their infant's needs and respond appropriately even while driving.

[0903] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0904] Step 1:

[0905] The user (parent) launches the smartphone app and uses the camera and microphone to collect the baby's facial expressions and cries.

[0906] Input: Video and audio data captured from the smartphone camera and microphone.

[0907] Output: Video and audio data of infants collected in real time.

[0908] Step 2:

[0909] The terminal transmits the collected video and audio data to a server.

[0910] Input: Real-time collected video and audio data of infants.

[0911] Output: Video and audio data of the infant sent to the server.

[0912] Step 3:

[0913] The server divides the received video data into frames and applies a face detection algorithm to extract the position and facial expression features of the infant's face.

[0914] Input: Video data sent to the server.

[0915] Output: Data containing face position and expression features.

[0916] Step 4:

[0917] The server extracts acoustic features (e.g., frequency spectrum, volume) from the received audio data.

[0918] Input: The audio data sent to the server.

[0919] Output: Data containing acoustic features.

[0920] Step 5:

[0921] The server analyzes the infant's needs using a generative AI model based on the extracted facial and acoustic features.

[0922] Input: Data including facial position, facial expression features, and acoustic features.

[0923] Output: Data identifying the infant's needs (e.g., hungry, sleepy, etc.).

[0924] Step 6:

[0925] The server generates a natural language message based on the infant's request and transmits it to the terminal.

[0926] Input: Data identifying the infant's needs.

[0927] Output: A natural language message such as "The baby is hungry. Please give him milk."

[0928] Step 7:

[0929] The device will display the received message on the parent's smartphone and notify the parent with a notification sound or vibration.

[0930] Input: A natural language message sent by the server.

[0931] Output: Messages and notifications that appear on your smartphone screen.

[0932] Step 8:

[0933] Parents follow the app's instructions to provide the necessary care for their infant, and then record the results (e.g., the infant's reaction after feeding) in the app.

[0934] Input: Data about parental behavior and its outcomes.

[0935] Output: Feedback data including the infant's reaction after feeding.

[0936] Step 9:

[0937] The terminal transmits the recorded feedback data to the server.

[0938] Input: Recorded feedback data.

[0939] Output: Feedback data sent to the server.

[0940] Step 10:

[0941] The server stores the feedback data in a database and uses it to retrain the AI ​​model.

[0942] Input: Feedback data.

[0943] Output: The updated AI model.

[0944] Step 11:

[0945] The server controls the driving of the car based on the results of the analysis. For example, if a baby is crying, the server can automatically slow down the car or instruct it to stop in a safe place.

[0946] Input: Notification results from analysis means.

[0947] Output: Car driving control instructions.

[0948] Step 12:

[0949] When an infant makes a request while driving, the server displays the request to the driver. This display is made on the vehicle's dashboard monitor or head-up display so that the driver can check it without looking away.

[0950] Input: Data about the infant's needs.

[0951] Output: Requests displayed on dashboard monitor or head-up display.

[0952] 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.

[0953] This invention is a system that analyzes an infant's facial expressions and cries to verbalize its requests and state, and further recognizes the emotional state of the user (parent) and adjusts the content of notifications. This system includes data collection means, data preprocessing means, analysis means, notification means, learning means, and an emotion engine.

[0954] Data collection methods

[0955] The data collection means include a camera and microphone to capture the infant's facial expressions and cries. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[0956] Data preprocessing measures

[0957] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is used to extract the position and facial expression of the infant's face. Acoustic features (e.g., frequency spectrum, volume) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[0958] Analysis means

[0959] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[0960] Notification means

[0961] The analysis results are converted into natural language on the server. For example, if the AI ​​model determines that the baby is hungry, it generates a specific message such as, "The baby is hungry. Please give him some milk." The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound. Furthermore, the content of the notification can be adjusted taking into account the user's emotional state.

[0962] Emotion Engine

[0963] An emotion engine is a device or software that analyzes a user's facial expressions and behavior to recognize their emotions. It identifies the user's stress level and emotional state (e.g., joy, anger, sadness). The emotion engine identifies the user's emotional state, and the notification mechanism adjusts the message to take this into account. For example, if the user is stressed, the notification content will include additional psychological support.

[0964] Learning tools

[0965] The user (parent) acts according to the advice and records the results in the app. For example, the device records the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[0966] Specific examples

[0967] The user (parent) launches a smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time, and this data is sent to a server. The server extracts facial features from the video data and acoustic features from the audio data. A generative AI model analyzes these features and identifies the baby's request as "hungry."

[0968] As a result, the server generates a message saying, "The baby is hungry. Please give him milk," and sends it to the device. The device then notifies the smartphone, and the user receives this notification and prepares the milk. The user then records the result in the app, and the device sends the feedback data to the server. The server uses this data to retrain the AI ​​model.

[0969] Furthermore, if the emotion engine analyzes the user's emotional state and determines that the user is feeling stressed, it will send a message containing psychological support such as, "The baby is hungry. Please feed him / her. Please stay calm and don't panic."

[0970] This system makes it possible to accurately grasp the needs of infants, reducing the burden of childcare on parents and providing appropriate support according to the user's emotional state.

[0971] The processing flow will be explained below.

[0972] Step 1:

[0973] The user (parent) launches the smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time.

[0974] Step 2:

[0975] The device (smartphone) collects video data taken with its camera and audio data recorded with its microphone. The collected data is temporarily stored on the device and then sent to a server.

[0976] Step 3:

[0977] The server receives the video data sent from the device, divides the video data into frames, and uses a face detection algorithm to identify the baby's face and extract its facial features (e.g., smiling, crying, frown, etc.).

[0978] Step 4:

[0979] The server simultaneously analyzes the received audio data and extracts acoustic features (e.g., frequency spectrum, volume, patterns, etc.). As a result, feature vectors are obtained from both the video and audio.

[0980] Step 5:

[0981] The server inputs the preprocessed data into the generative AI model, which uses facial and acoustic features to refer to past learning data to identify the infant's needs (e.g., "I'm hungry" or "I'm sleepy") and obtains the analysis results.

[0982] Step 6:

[0983] The server converts the analysis results into natural language. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him milk."

[0984] Step 7:

[0985] The server sends the generated message to the device, which receives it and notifies the user's smartphone.

[0986] Step 8:

[0987] To enable the device to recognize the user's emotional state, the device uses the smartphone's front camera to collect facial expression data, which is then sent to a server.

[0988] Step 9:

[0989] The server uses an emotion engine to analyze the user's facial expression data, which determines the user's stress level and emotional state (e.g., happy, anger, sadness).

[0990] Step 10:

[0991] The server adjusts the notification content based on the user's emotional state. For example, if the user is under stress, the server generates a message containing psychological support such as, "The baby is hungry. Please feed him / her. Please stay calm and don't panic."

[0992] Step 11:

[0993] The server sends the adjusted notification message to the device, which then notifies the user via a screen display, notification sound, or vibration.

[0994] Step 12:

[0995] The user receives a notification and acts on the advice, for example, preparing milk and feeding the baby, and then records the results in the app (for example, the baby's reaction after feeding).

[0996] Step 13:

[0997] The terminal transmits feedback data from the user to the server, including the user's behavior and the infant's reactions.

[0998] Step 14:

[0999] The server stores the received feedback data in a database, which is then periodically used to retrain the AI ​​model, improving the analysis accuracy of the AI ​​model and the overall performance of the system.

[1000] Example 2

[1001] 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."

[1002] Accurately understanding an infant's needs is a significant burden for many parents and often causes parenting stress. Furthermore, depending on the parent's emotional state, they may not be able to respond appropriately. To solve these problems, there is a need for a system that analyzes an infant's facial expressions and cries, verbalizes their needs, and recognizes the parent's emotional state to adjust the content of notifications accordingly.

[1003] 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.

[1004] In this invention, the server includes a data collection means for collecting the facial expressions and cries of the infant, a data preprocessing means for preprocessing the collected facial expression and cries data, an analysis means for analyzing the preprocessed data to identify the infant's needs, a notification means for converting the analysis results into natural language and notifying the parent, a learning means for recording the infant's responses and relearning based on the results, and an emotion engine for recognizing the user's emotional state and adjusting the content of the notification. This makes it possible to accurately grasp the infant's needs and to provide appropriate notifications according to the parent's emotional state.

[1005] "Infant facial expressions and vocalizations" refers to the patterns of an infant's facial movements, facial expressions, and oral sounds.

[1006] "Data collection means" refers to devices or systems for collecting the infant's facial expressions and cries. Specifically, cameras and microphones are used.

[1007] "Data pre-processing methods" refer to techniques and processes used to convert collected raw data into a form suitable for analysis, such as video frame segmentation or acoustic feature extraction.

[1008] "Analytical tools" refer to algorithms and techniques used to identify infant needs based on pre-processed data, specifically generative AI models.

[1009] "Notification means" refers to the technology or system that converts the analysis results into natural language and notifies the parents, for example by displaying a message on a smartphone.

[1010] "Learning methods" refer to the techniques and methods used to record the infant's responses and re-learn based on those results. Specifically, the feedback data is used to improve the accuracy of the AI ​​model.

[1011] An "emotion engine" is a technology or system that recognizes a user's emotional state and adjusts notification content accordingly, for example by analyzing the user's facial expressions and voice.

[1012] MODE FOR CARRYING OUT THE INVENTION

[1013] This invention is a system that analyzes an infant's facial expressions and cries to verbalize its requests and state, and further recognizes the emotional state of the user (parent) and adjusts the content of notifications. This system includes data collection means, data preprocessing means, analysis means, notification means, learning means, and an emotion engine.

[1014] Data collection methods

[1015] Users use smartphones or baby monitors to collect video and audio of their infants in real time. This is achieved by using the smartphone's camera and microphone as data collection means. Through this operation, the infant's movements and cries are stored as digital data on the device.

[1016] Data preprocessing measures

[1017] The device sends the collected video and audio data to a server. The server divides the video data into frames and uses a face detection algorithm to extract the position and facial expression features of the infant's face. Acoustic features (such as frequency spectrum and sound intensity) are also extracted from the audio data. Libraries such as OpenCV, Dlib, and Librosa are used for this preprocessing. Preprocessing allows the data to be formatted before being input into the AI ​​model.

[1018] Analysis means

[1019] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if the facial and acoustic features match the pattern of "hungry," the AI ​​model will determine that the infant is "hungry." This analysis is performed using deep learning frameworks such as TensorFlow and PyTorch.

[1020] Notification means

[1021] The analysis results are converted into natural language on the server. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him some milk." The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is displayed on the smartphone screen and uses a notification sound, and is sent via Firebase Cloud Messaging or Apple Push Notification Service.

[1022] Emotion Engine

[1023] An emotion engine is a device or software that analyzes a user's facial expressions and behavior to recognize their emotional state. It uses emotion recognition libraries such as TensorFlow and OpenFace. For example, it identifies the user's stress level and emotional state (e.g., joy, anger, sadness). The emotion engine identifies the user's emotional state, and the notification method adjusts the message to take this into account. For example, if the user is stressed, an additional message such as "Please remain calm and respond calmly" is included.

[1024] Learning tools

[1025] The user acts according to the notification and records the results in the app. For example, the user can record the baby's reaction after feeding the baby. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the generative AI model, improving its analytical accuracy.

[1026] Specific examples

[1027] The user launches a smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time, and these data are sent to a server. The server extracts facial features from the video data and acoustic features from the audio data. A generative AI model analyzes these features and identifies the baby's request as "hungry." As a result, the server generates a message saying, "The baby is hungry. Please feed him / her." and sends it to the device. The device notifies the smartphone, and the user prepares the milk. The user then records the result in the app, and the device sends feedback data to the server. The server uses this data to retrain the generative AI model. Furthermore, an emotion engine analyzes the user's emotional state. If it determines that the user is stressed, it sends a message containing psychological support, such as, "The baby is hungry. Please feed him / her. Please stay calm and respond calmly."

[1028] Prompt Sentence Examples

[1029] "Please explain the data you would like to input into a generative AI model for a system that analyzes a baby's facial expressions and vocalizations to determine its needs and notify the parents."

[1030] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1031] Step 1:

[1032] The user launches the smartphone app and points the camera at the baby. When the user starts recording with the app, the smartphone's camera and microphone begin collecting video and audio of the baby. The input at this stage is the baby's real-time facial expressions and cries, and the output is video and audio data stored on the smartphone as digital data.

[1033] Step 2:

[1034] The smartphone (device) sends the collected video and audio data to a server using a communication method, specifically Wi-Fi or mobile data communication. The input at this stage is the video and audio data stored on the smartphone, and the output is digital data transferred to the server.

[1035] Step 3:

[1036] The server divides the received video data into frames and extracts the infant's facial position and facial expression features using a face detection algorithm (e.g., OpenCV or Dlib). It also extracts acoustic features (e.g., frequency spectrum and sound intensity) from the audio data using an audio processing library such as Librosa. The input at this stage is the video and audio data sent to the server, and the output is the processed facial and acoustic features.

[1037] Step 4:

[1038] The server inputs the preprocessed facial and acoustic features into a generative AI model to analyze the infant's needs. Specifically, the model performs analysis using deep learning frameworks such as TensorFlow and PyTorch to identify the infant's needs (e.g., "I'm hungry"). The input at this stage is the preprocessed feature data, and the output is the analyzed infant's needs.

[1039] Step 5:

[1040] The server converts the analysis results into natural language. This is achieved using an NLP engine such as GPT-3 or BERT. For example, if the analysis result is "hungry," the server generates a message saying "The baby is hungry. Please give him milk." The input at this stage is the analysis result as the infant's request, and the output is a notification message in natural language.

[1041] Step 6:

[1042] The server then sends the generated notification message to the device (smartphone). This is done via Firebase Cloud Messaging (FCM) or Apple Push Notification Service (APNs). The user receives the message in real time on their smartphone and is notified by a notification sound or a display on the screen. The input at this stage is a natural language notification message, and the output is a notification displayed on the smartphone.

[1043] Step 7:

[1044] The server records the user's facial expressions and behavior and analyzes their emotional state using an emotion engine (such as TensorFlow or OpenFace). If the server determines that the user is feeling stressed, it includes an additional psychological support message in the notification. The input of this stage is the user's facial expressions and voice data, and the output is a tailored notification message.

[1045] Step 8:

[1046] The user acts according to the notification and records the results in the app. For example, the user can enter the baby's reaction after feeding into the app. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the generative AI model. The input at this stage is the feedback data recorded by the user in the app, and the output is the retraining data stored on the server.

[1047] (Application example 2)

[1048] 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."

[1049] A system is needed that can not only accurately grasp the needs and status of infants and notify parents in real time, but also recognize the parents' emotional state and provide appropriate notifications according to the situation. In particular, there is a need for a system that can respond quickly to infants' needs while taking into consideration the parents' stress levels and emotional state, thereby reducing the burden of childcare and providing higher quality childcare support.

[1050] 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.

[1051] In this invention, the server includes data collection means for collecting the facial expressions and cries of the infant, data preprocessing means for preprocessing the collected facial expression and cries data, analysis means for analyzing the preprocessed data to identify the infant's needs, notification means for converting the analysis results into natural language and notifying the guardian, learning means for recording the infant's response results and relearning based on those results, and an emotion engine for recognizing the user's emotional state and adjusting the content of the notification. This makes it possible to accurately grasp the infant's needs and provide appropriate support according to the guardian's emotional state.

[1052] The "data collection means" is a device that includes a camera and a microphone to collect the infant's facial expressions and sounds.

[1053] The "data preprocessing means" is a means for executing a process to prepare the collected facial expression and vocalization data in a form suitable for analysis.

[1054] "Analysis means" means a means of using a generative AI model to identify the needs of an infant based on pre-processed data.

[1055] The "notification means" is a means for converting the analysis results into natural language and notifying the parents.

[1056] The "learning means" is a means for recording the infant's response results and using that information to improve the system's analytical accuracy.

[1057] The "emotion engine" is a means of recognizing the parent's emotional state and adjusting the content of notifications based on that information.

[1058] A "generative AI model" is an artificial intelligence model that identifies the needs and conditions of infants and young children based on collected and pre-processed data.

[1059] The system for implementing this invention collects and analyzes the facial expressions and cries of infants, and then notifies parents appropriately, thereby reducing the burden of childcare. It also recognizes the parents' emotional state and adjusts the content of notifications based on that information. This system utilizes various hardware and software and is implemented as follows:

[1060] Hardware and software used

[1061] Hardware: smartphone, smart glasses, server, camera and microphone.

[1062] Software: Data collection application (for smartphone / smart glasses), data pre-processing algorithm (server), emotion recognition model (generative AI model), notification application (smartphone).

[1063] System Overview

[1064] Data collection methods

[1065] The system uses a device (such as a smartphone or smart glasses) to collect the infant's facial expressions and cries. This uses a camera and microphone, and the collected video and audio data is sent to a server via the internet.

[1066] Data preprocessing measures

[1067] The server splits the video data into frames and performs face detection using the OpenCV library. It also extracts acoustic features such as frequency spectrum and volume from the audio data. These preprocessing steps prepare the data in a format suitable for analysis.

[1068] Analysis means

[1069] Based on the pre-processed data, a generative AI model is used to identify the infant's needs, such as whether they are hungry or sleepy. This analysis also utilizes past training data, so accuracy improves over time.

[1070] Notification means

[1071] The analysis results are converted into natural language on the server and sent to the parent's device. Notifications are displayed on the smartphone screen or via voice notification. The emotion engine also recognizes the parent's emotional state and adjusts the content of the notification based on that information. For example, if the parent is feeling stressed, a message containing psychological support will be sent.

[1072] Learning tools

[1073] Parents act on the advice provided and record their results in the app. This feedback data is sent to the server and used to retrain the model, further improving its analysis accuracy.

[1074] Specific example explanation

[1075] 1. Data Collection

[1076] Parents launch the smartphone app and take a picture of their baby's face with the camera. Audio data is also collected, and this data is sent to a server in real time.

[1077] 2. Data Preprocessing

[1078] The server preprocesses the data that arrives, extracting facial features from video data and acoustic features from audio data.

[1079] 3. Analysis

[1080] A generative AI model is used to analyze an infant's needs, for example, to identify the need to "be hungry."

[1081] 4. Notification

[1082] A message is generated and sent to the parent's smartphone saying, "Your baby is hungry. Please feed him / her." If the parent's emotional state indicates stress, an additional message is included saying, "Please stay calm."

[1083] Prompt Sentence Examples

[1084] Prompt: "Create a system that detects delivery workers' fatigue and suggests breaks. Collect facial expression data from the delivery workers using a camera and analyze the fatigue indicators using a model."

[1085] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1086] Step 1: Data collection

[1087] The user (guardian) starts the smartphone app and takes a picture of the baby's face with the camera. At the same time, audio is collected with the microphone. The collected video data (facial expressions) and audio data (cries) are sent from the device to the server in real time. This makes the input data available on the server.

[1088] Step 2: Data Preprocessing

[1089] The server splits the received video data into frames using the OpenCV library and performs face detection, extracting facial features. At the same time, acoustic features such as frequency spectrum and volume are extracted from the audio data. Once preprocessing is complete, these feature data become available as input data for analysis.

[1090] Step 3: Data analysis

[1091] The server inputs the preprocessed facial and acoustic features into a generative AI model. The generative AI model analyzes this data and identifies the infant's needs. For example, the request "I'm hungry" may be identified. The analysis results are output as a natural language message indicating a specific response.

[1092] Step 4: Notification

[1093] The server then sends the generated natural language message to the parent's smartphone. Notifications are displayed on the screen or sounded. The emotion engine then analyzes the parent's emotional state, and if it detects stress, it adds psychological support to the message. For example, it may include a message saying, "Please stay calm."

[1094] Step 5: Gather feedback

[1095] The user (guardian) takes action according to the notification and then records the results in the smartphone app. For example, they input "the baby's reaction after giving milk." This feedback data is sent from the device to the server.

[1096] Step 6: Retrain the model

[1097] The server saves the collected feedback data in a database and uses it to retrain the generative AI model, thereby improving the accuracy of analysis. As the feedback data accumulated sequentially is used as training data for the model, the accuracy of the model improves over time.

[1098] Based on this processing step, the system can accurately grasp the needs of infants and provide appropriate notifications that take into account the emotional state of the caregiver, thereby reducing the burden of childcare on the caregiver.

[1099] 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.

[1100] 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.

[1101] 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.

[1102] [Fourth embodiment]

[1103] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1104] 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.

[1105] 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).

[1106] 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.

[1107] 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.

[1108] 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).

[1109] 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.

[1110] 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.

[1111] 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.

[1112] 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.

[1113] 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.

[1114] 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.

[1115] 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."

[1116] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, and a learning means.

[1117] Data collection methods

[1118] The data collection means include a camera and microphone to capture the infant's facial expressions and cries. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[1119] Data preprocessing measures

[1120] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is applied to extract the position and facial expression features of the infant's face. Acoustic features (e.g., frequency spectrum, sound intensity) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[1121] Analysis means

[1122] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[1123] Notification means

[1124] The analysis results are converted into natural language on the server. For example, a specific message such as "The baby is hungry. Please give him milk" is generated. The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound.

[1125] Learning tools

[1126] The user (parent) acts according to the advice and records the results in the app. For example, the device records the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[1127] Specific examples

[1128] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[1129] This system will make it possible to accurately grasp an infant's needs, and is expected to reduce the burden of childcare on parents.

[1130] The processing flow will be explained below.

[1131] Step 1:

[1132] The user (parent) launches the smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time.

[1133] Step 2:

[1134] The device (smartphone) collects video data taken with the camera and audio data recorded with the microphone. This data is temporarily stored on the device and immediately sent to the server.

[1135] Step 3:

[1136] The server receives the video data sent from the device. The video data is divided into frames, and a face detection algorithm is used to identify the baby's face and extract its facial features (e.g., smiling, crying, frown, etc.).

[1137] Step 4:

[1138] The server simultaneously analyzes the received audio data and extracts acoustic features (e.g., frequency spectrum, volume, patterns, etc.). As a result, feature vectors are obtained from both the video and audio.

[1139] Step 5:

[1140] The server inputs this preprocessed data into a generative AI model, which uses facial and acoustic features to identify the infant's needs (e.g., "I'm hungry" or "I'm sleepy") by referring to past learning data.

[1141] Step 6:

[1142] The server converts the analysis results into natural language. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him milk."

[1143] Step 7:

[1144] The server generates a message and sends it to the device, which receives it and notifies the user's smartphone. The notification can be displayed on the screen, sounded, or vibrated to inform the parent.

[1145] Step 8:

[1146] The user receives a notification and acts on the advice, for example, preparing milk and feeding the baby, and then records the results in the app (e.g., the baby's reaction after feeding).

[1147] Step 9:

[1148] The device sends user feedback data to the server, detailing the infant's condition and response results.

[1149] Step 10:

[1150] The server stores the received feedback data in a database. The stored data is periodically used to retrain the AI ​​model, improving the analytical accuracy of the AI ​​model and enabling more accurate judgments.

[1151] Example 1

[1152] 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."

[1153] It is difficult for parents to quickly and accurately grasp their infant's needs and state, which increases the burden of childcare. Furthermore, there is a lack of means to identify appropriate needs from an infant's facial expressions and cries, which can lead to stress due to delayed or incorrect responses. Therefore, a system is needed that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized information to parents.

[1154] 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.

[1155] In this invention, the server includes a data collection means for collecting the facial expressions and cries of the infant, a data preprocessing means for preprocessing the collected facial expression and cries data and extracting the position of the face and acoustic features, an analysis means for identifying the infant's needs using a generative AI model based on the preprocessed data, a notification means for converting the analysis results into natural language and notifying the parent in real time, and a learning means for recording the infant's response results and relearning the generative AI model based on the results. This makes it possible to appropriately and quickly grasp the infant's needs and condition and reduce the burden of childcare on parents.

[1156] The "data collection means" is a device that uses a camera and a microphone to collect the infant's facial expressions and cries and obtain them as digital data.

[1157] The "data preprocessing means" is a process that preprocesses the collected facial expression and sound data, divides the video data into frames to extract facial position and facial expression features, and extracts acoustic features from the audio data.

[1158] The "analysis means" is an algorithm that uses a generative AI model based on pre-processed data to identify the infant's needs and classify them based on facial and acoustic features.

[1159] The "notification means" is a system that converts the analysis results into natural language and notifies parents in real time, and is a function that sends messages via the smartphone screen display, notification sound, etc.

[1160] The "learning method" is the process of recording the infant's response results and relearning the generative AI model based on those results, thereby improving the accuracy of the analysis.

[1161] A "generative AI model" is an artificial intelligence algorithm used to identify an infant's needs using collected and pre-processed data.

[1162] "Facial expression features" are data that indicate changes in facial position and facial expression extracted from video data.

[1163] "Acoustic features" are data that indicate the frequency spectrum, sound intensity, etc. extracted from the audio data.

[1164] "Natural language" refers to the language that humans use on a daily basis, and is the language used to communicate the analysis results to parents in a form that is easy to understand.

[1165] "Relearning" is the process of updating the artificial intelligence algorithm based on collected feedback data to improve analysis accuracy.

[1166] "Real-time" means that the process from data collection to analysis and notification is carried out immediately without delay.

[1167] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, and a learning means.

[1168] Data collection methods

[1169] Users use smartphones or baby monitors to collect real-time video and audio of their infants. This involves using a camera and microphone to capture the infant's movements and cries as digital data. This allows the collection of various facial expressions and cries of the infant, which can then be used for subsequent processing.

[1170] Data preprocessing measures

[1171] The collected digital data is sent from the device (smartphone) to a server. The server first divides the video data into frames and uses a face detection algorithm to extract the position of the infant's face and facial expression features. For example, a library such as OpenCV is used to perform real-time face detection and extract facial expression data. Meanwhile, an FFT (fast Fourier transform) is used on the audio data to extract acoustic features such as frequency spectrum and sound intensity. These preprocessing steps shape the input data for the generative AI model.

[1172] Analysis means

[1173] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. This analysis is performed using machine learning frameworks such as PyTorch. The AI ​​model uses facial and acoustic features as input to classify the infant's needs, such as "hungry" or "sleepy." The accuracy of the analysis can be improved by utilizing past training data.

[1174] Notification means

[1175] The analysis results are converted into natural language on the server. For example, a message such as "The baby is hungry. Please give him milk" is generated. This generated message is sent from the server to the device (smartphone) and notified to the user in real time via the smartphone's screen display and notification sound.

[1176] Learning tools

[1177] The user (parent) acts according to the advice and records the results in the app. For example, they record the baby's reaction after feeding. This data is sent to the server via the device, and the server stores it in a database. The stored data is used to retrain the generative AI model, improving the model's analytical accuracy.

[1178] Specific examples

[1179] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[1180] Prompt Sentence Examples

[1181] "When your baby starts crying, open the smartphone app and use the camera and microphone to collect facial expressions and sounds. Press the request button, and the AI ​​model will provide appropriate advice based on the analysis results. After the analysis, you can record and send the behavioral results to the app, which will help improve the accuracy of the AI."

[1182] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1183] Step 1: Data collection

[1184] Server initialization of data collection methods

[1185] Input: The user launches the smartphone app and enables the camera and microphone.

[1186] Specific operation: The user launches the smartphone app and presses the start recording button.

[1187] Output: Real-time video and audio data is collected on a smartphone.

[1188] Step 2: Send data

[1189] Sending data to the server by the device

[1190] Input: Collected video and audio data.

[1191] Specific operation: The terminal divides the collected data into packets and sends them to a server via the Internet.

[1192] Output: Digital data (video and audio data) that arrives at the server.

[1193] Step 3: Data Preprocessing (video)

[1194] Preprocessing of video data by the server

[1195] Input: Video data arriving at the server.

[1196] Specific operation: The server divides the video data into frames and extracts the position and facial expression features of the infant's face using a face detection algorithm (e.g., OpenCV).

[1197] Output: Face position and facial expression feature data for each frame.

[1198] Step 4: Data Preprocessing (Audio)

[1199] Preprocessing of audio data by the server

[1200] Input: The audio data arriving at the server.

[1201] Specific operation: The server applies FFT (Fast Fourier Transform) to the audio data to extract the frequency spectrum and sound intensity.

[1202] Output: Acoustic features of the audio data (frequency spectrum, sound intensity).

[1203] Step 5: Data analysis

[1204] Analysis using AI models generated by the server

[1205] Input: Extracted facial and acoustic feature data.

[1206] How it works: The server inputs the preprocessed data into a generative AI model (e.g., built using PyTorch) to analyze the infant's needs (e.g., "I'm hungry," "I'm sleepy," etc.).

[1207] Output: A request classification result, such as "I'm hungry."

[1208] Step 6: Generate a notification message

[1209] Natural language generation of analysis results by the server

[1210] Input: Requirements classification results from the generative AI model.

[1211] Specific operation: The server uses NLP (natural language processing) technology to convert the request classification results into a message that is easy for the parent to understand (e.g., "The baby is hungry. Please give him milk.").

[1212] Output: A naturalized message.

[1213] Step 7: Notification

[1214] Server sends message to device

[1215] Input: A naturalized message.

[1216] Specific operation: The server uses push notification to send the generated message to the terminal.

[1217] Output: The message arrives at the terminal and the user is notified by the smartphone screen display and notification sound.

[1218] Step 8: User response and feedback

[1219] User response and input of results

[1220] Input: The user acts on the notification (e.g., gives milk) and records the result in the app (e.g., "reaction after giving milk").

[1221] Specific operation: The user enters the corresponding results into the application form and presses the submit button.

[1222] Output: Feedback data recorded on the device.

[1223] Step 9: Sending feedback to the server

[1224] Sending feedback data from the device to the server

[1225] Input: Feedback data recorded on the device.

[1226] Specific operation: The terminal divides the feedback data into packets and sends them to the server via the Internet.

[1227] Output: The feedback data that arrives at the server.

[1228] Step 10: Retrain

[1229] Retraining of the generated AI model by the server

[1230] Input: Collected feedback data.

[1231] Specific operation: The server periodically ingests feedback data in batches and retrains the generative AI model.

[1232] Output: Improved analysis accuracy with updated AI model.

[1233] (Application example 1)

[1234] 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."

[1235] When driving a car with an infant, parents are required to respond quickly to their infant's requests while concentrating on driving. However, it is difficult to check the infant's facial expressions or cries while driving, which can result in delayed responses. Furthermore, if parents are unable to understand and respond appropriately to their infant's requests while driving, the infant's safety and comfort may be compromised. A system is needed to improve this situation and respond quickly and accurately to infant requests even while driving.

[1236] 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.

[1237] In this invention, the server includes a data collection means, a data preprocessing means, an analysis means, a notification means, a learning means, a control means for controlling the driving of the vehicle based on the notification results from the analysis means, and a display means for displaying the content of a request to the driver when a request from the infant occurs while driving. This allows parents to quickly understand the infant's request and respond appropriately even while driving.

[1238] The "data collection means" is a device that includes a camera and a microphone to collect the infant's facial expressions and sounds.

[1239] The "data pre-processing means" is a processing device for converting the collected facial expression and vocalization data into a format suitable for analysis.

[1240] The "analysis means" is a function for identifying infants' needs based on pre-processed data, and utilizes a generative AI model.

[1241] The "notification means" is a function for converting the analysis results into natural language and notifying the parents.

[1242] The "learning means" is a function that records the infant's response results and re-learns the system based on those results.

[1243] The "control means" is a device for controlling the driving of the vehicle based on the results of notification by the analysis means.

[1244] The "display means" is a device for displaying the request content to the driver when a request is made by the baby while driving.

[1245] This invention is a system that analyzes an infant's facial expressions and cries, verbalizes the infant's needs and state, and provides the verbalized results to the parent. This system includes a data collection means, a data preprocessing means, an analysis means, a notification means, a learning means, a control means, and a display means.

[1246] Data collection methods

[1247] The data collection means includes a camera and microphone to capture the facial expressions and cries of the infant. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[1248] Data preprocessing measures

[1249] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is applied to extract the position and facial expression features of the infant's face. Acoustic features (e.g., frequency spectrum, volume) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[1250] Analysis means

[1251] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[1252] Notification means

[1253] The analysis results are converted into natural language on the server. For example, a specific message such as "The baby is hungry. Please give him milk" is generated. The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound.

[1254] Learning tools

[1255] The user (parent) acts according to the advice and records the results in the app. For example, recording the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[1256] Control means

[1257] The server controls the driving of the car based on the results of the analysis. For example, if a baby is crying, the server can automatically slow down the car or instruct it to stop in a safe place.

[1258] Display means

[1259] If a request is made by an infant while driving, the server displays the request to the driver, using the vehicle's dashboard monitor or head-up display so that the driver can check it without looking away.

[1260] Specific examples

[1261] When the baby starts crying in the morning, the user (parent) launches the smartphone app. The camera and microphone are activated, collecting the baby's facial expressions and cries in real time. The device sends this data to the server, which extracts facial and acoustic features. The AI ​​model analyzes the data and determines that the baby is hungry, generating a message saying, "The baby is hungry. Please give him / her milk." The device then notifies the smartphone of this message, and the user quickly prepares the milk. This series of actions allows the user to respond to the baby's needs accurately and quickly. The user then records the situation after giving him / her milk in the app and sends it to the server. The server uses this data to retrain the AI ​​model, improving the accuracy of future analyses.

[1262] An example of a prompt statement is, "What should you do if your baby is crying?"

[1263] This system allows parents to quickly and accurately grasp their infant's needs and respond appropriately even while driving.

[1264] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1265] Step 1:

[1266] The user (parent) launches the smartphone app and uses the camera and microphone to collect the baby's facial expressions and cries.

[1267] Input: Video and audio data captured from the smartphone camera and microphone.

[1268] Output: Video and audio data of infants collected in real time.

[1269] Step 2:

[1270] The terminal transmits the collected video and audio data to a server.

[1271] Input: Real-time collected video and audio data of infants.

[1272] Output: Video and audio data of the infant sent to the server.

[1273] Step 3:

[1274] The server divides the received video data into frames and applies a face detection algorithm to extract the position and facial expression features of the infant's face.

[1275] Input: Video data sent to the server.

[1276] Output: Data containing face position and expression features.

[1277] Step 4:

[1278] The server extracts acoustic features (e.g., frequency spectrum, volume) from the received audio data.

[1279] Input: The audio data sent to the server.

[1280] Output: Data containing acoustic features.

[1281] Step 5:

[1282] The server analyzes the infant's needs using a generative AI model based on the extracted facial and acoustic features.

[1283] Input: Data including facial position, facial expression features, and acoustic features.

[1284] Output: Data identifying the infant's needs (e.g., hungry, sleepy, etc.).

[1285] Step 6:

[1286] The server generates a natural language message based on the infant's request and transmits it to the terminal.

[1287] Input: Data identifying the infant's needs.

[1288] Output: A natural language message such as "The baby is hungry. Please give him milk."

[1289] Step 7:

[1290] The device will display the received message on the parent's smartphone and notify the parent with a notification sound or vibration.

[1291] Input: A natural language message sent by the server.

[1292] Output: Messages and notifications that appear on your smartphone screen.

[1293] Step 8:

[1294] Parents follow the app's instructions to provide the necessary care for their infant, and then record the results (e.g., the infant's reaction after feeding) in the app.

[1295] Input: Data about parental behavior and its outcomes.

[1296] Output: Feedback data including the infant's reaction after feeding.

[1297] Step 9:

[1298] The terminal transmits the recorded feedback data to the server.

[1299] Input: Recorded feedback data.

[1300] Output: Feedback data sent to the server.

[1301] Step 10:

[1302] The server stores the feedback data in a database and uses it to retrain the AI ​​model.

[1303] Input: Feedback data.

[1304] Output: The updated AI model.

[1305] Step 11:

[1306] The server controls the driving of the car based on the results of the analysis. For example, if a baby is crying, the server can automatically slow down the car or instruct it to stop in a safe place.

[1307] Input: Notification results from analysis means.

[1308] Output: Car driving control instructions.

[1309] Step 12:

[1310] When an infant makes a request while driving, the server displays the request to the driver. This display is made on the vehicle's dashboard monitor or head-up display so that the driver can check it without looking away.

[1311] Input: Data about the infant's needs.

[1312] Output: Requests displayed on dashboard monitor or head-up display.

[1313] 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.

[1314] This invention is a system that analyzes an infant's facial expressions and cries to verbalize its requests and state, and further recognizes the emotional state of the user (parent) and adjusts the content of notifications. This system includes data collection means, data preprocessing means, analysis means, notification means, learning means, and an emotion engine.

[1315] Data collection methods

[1316] The data collection means include a camera and microphone to capture the infant's facial expressions and cries. Users (parents) use smartphones or baby monitors to collect video and audio of the infant in real time, allowing the system to capture the infant's movements and cries as digital data.

[1317] Data preprocessing measures

[1318] Data sent from the device (smartphone) is preprocessed on the server. The video data is divided into frames, and a face detection algorithm is used to extract the position and facial expression of the infant's face. Acoustic features (e.g., frequency spectrum, volume) are extracted from the audio data. These preprocessing steps shape the input data for the AI ​​model.

[1319] Analysis means

[1320] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if facial and acoustic features are identified as contributing factors to "hungry," the AI ​​model will determine that the infant is "hungry." This analysis utilizes past learning data to make highly accurate judgments.

[1321] Notification means

[1322] The analysis results are converted into natural language on the server. For example, if the AI ​​model determines that the baby is hungry, it generates a specific message such as, "The baby is hungry. Please give him some milk." The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is notified to the user by a display on the smartphone screen or a notification sound. Furthermore, the content of the notification can be adjusted taking into account the user's emotional state.

[1323] Emotion Engine

[1324] An emotion engine is a device or software that analyzes a user's facial expressions and behavior to recognize their emotions. It identifies the user's stress level and emotional state (e.g., joy, anger, sadness). The emotion engine identifies the user's emotional state, and the notification mechanism adjusts the message to take this into account. For example, if the user is stressed, the notification content will include additional psychological support.

[1325] Learning tools

[1326] The user (parent) acts according to the advice and records the results in the app. For example, the device records the baby's reaction after feeding. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the AI ​​model, improving its analytical accuracy.

[1327] Specific examples

[1328] The user (parent) launches a smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time, and this data is sent to a server. The server extracts facial features from the video data and acoustic features from the audio data. A generative AI model analyzes these features and identifies the baby's request as "hungry."

[1329] As a result, the server generates a message saying, "The baby is hungry. Please give him milk," and sends it to the device. The device then notifies the smartphone, and the user receives this notification and prepares the milk. The user then records the result in the app, and the device sends the feedback data to the server. The server uses this data to retrain the AI ​​model.

[1330] Furthermore, if the emotion engine analyzes the user's emotional state and determines that the user is feeling stressed, it will send a message containing psychological support such as, "The baby is hungry. Please feed him / her. Please stay calm and don't panic."

[1331] This system makes it possible to accurately grasp the needs of infants, reducing the burden of childcare on parents and providing appropriate support according to the user's emotional state.

[1332] The processing flow will be explained below.

[1333] Step 1:

[1334] The user (parent) launches the smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time.

[1335] Step 2:

[1336] The device (smartphone) collects video data taken with its camera and audio data recorded with its microphone. The collected data is temporarily stored on the device and then sent to a server.

[1337] Step 3:

[1338] The server receives the video data sent from the device, divides the video data into frames, and uses a face detection algorithm to identify the baby's face and extract its facial features (e.g., smiling, crying, frown, etc.).

[1339] Step 4:

[1340] The server simultaneously analyzes the received audio data and extracts acoustic features (e.g., frequency spectrum, volume, patterns, etc.). As a result, feature vectors are obtained from both the video and audio.

[1341] Step 5:

[1342] The server inputs the preprocessed data into the generative AI model, which uses facial and acoustic features to refer to past learning data to identify the infant's needs (e.g., "I'm hungry" or "I'm sleepy") and obtains the analysis results.

[1343] Step 6:

[1344] The server converts the analysis results into natural language. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him milk."

[1345] Step 7:

[1346] The server sends the generated message to the device, which receives it and notifies the user's smartphone.

[1347] Step 8:

[1348] To enable the device to recognize the user's emotional state, the device uses the smartphone's front camera to collect facial expression data, which is then sent to a server.

[1349] Step 9:

[1350] The server uses an emotion engine to analyze the user's facial expression data, which determines the user's stress level and emotional state (e.g., happy, anger, sadness).

[1351] Step 10:

[1352] The server adjusts the notification content based on the user's emotional state. For example, if the user is under stress, the server generates a message containing psychological support such as, "The baby is hungry. Please feed him / her. Please stay calm and don't panic."

[1353] Step 11:

[1354] The server sends the adjusted notification message to the device, which then notifies the user via a screen display, notification sound, or vibration.

[1355] Step 12:

[1356] The user receives a notification and acts on the advice, for example, preparing milk and feeding the baby, and then records the results in the app (for example, the baby's reaction after feeding).

[1357] Step 13:

[1358] The terminal transmits feedback data from the user to the server, including the user's behavior and the infant's reactions.

[1359] Step 14:

[1360] The server stores the received feedback data in a database, which is then periodically used to retrain the AI ​​model, improving the analysis accuracy of the AI ​​model and the overall performance of the system.

[1361] Example 2

[1362] 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."

[1363] Accurately understanding an infant's needs is a significant burden for many parents and often causes parenting stress. Furthermore, depending on the parent's emotional state, they may not be able to respond appropriately. To solve these problems, there is a need for a system that analyzes an infant's facial expressions and cries, verbalizes their needs, and recognizes the parent's emotional state to adjust the content of notifications accordingly.

[1364] 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.

[1365] In this invention, the server includes a data collection means for collecting the facial expressions and cries of the infant, a data preprocessing means for preprocessing the collected facial expression and cries data, an analysis means for analyzing the preprocessed data to identify the infant's needs, a notification means for converting the analysis results into natural language and notifying the parent, a learning means for recording the infant's responses and relearning based on the results, and an emotion engine for recognizing the user's emotional state and adjusting the content of the notification. This makes it possible to accurately grasp the infant's needs and to provide appropriate notifications according to the parent's emotional state.

[1366] "Infant facial expressions and vocalizations" refers to the patterns of an infant's facial movements, facial expressions, and oral sounds.

[1367] "Data collection means" refers to devices or systems for collecting the infant's facial expressions and cries. Specifically, cameras and microphones are used.

[1368] "Data pre-processing methods" refer to techniques and processes used to convert collected raw data into a form suitable for analysis, such as video frame segmentation or acoustic feature extraction.

[1369] "Analytical tools" refer to algorithms and techniques used to identify infant needs based on pre-processed data, specifically generative AI models.

[1370] "Notification means" refers to the technology or system that converts the analysis results into natural language and notifies the parents, for example by displaying a message on a smartphone.

[1371] "Learning methods" refer to the techniques and methods used to record the infant's responses and re-learn based on those results. Specifically, the feedback data is used to improve the accuracy of the AI ​​model.

[1372] An "emotion engine" is a technology or system that recognizes a user's emotional state and adjusts notification content accordingly, for example by analyzing the user's facial expressions and voice.

[1373] MODE FOR CARRYING OUT THE INVENTION

[1374] This invention is a system that analyzes an infant's facial expressions and cries to verbalize its requests and state, and further recognizes the emotional state of the user (parent) and adjusts the content of notifications. This system includes data collection means, data preprocessing means, analysis means, notification means, learning means, and an emotion engine.

[1375] Data collection methods

[1376] Users use smartphones or baby monitors to collect video and audio of their infants in real time. This is achieved by using the smartphone's camera and microphone as data collection means. Through this operation, the infant's movements and cries are stored as digital data on the device.

[1377] Data preprocessing measures

[1378] The device sends the collected video and audio data to a server. The server divides the video data into frames and uses a face detection algorithm to extract the position and facial expression features of the infant's face. Acoustic features (such as frequency spectrum and sound intensity) are also extracted from the audio data. Libraries such as OpenCV, Dlib, and Librosa are used for this preprocessing. Preprocessing allows the data to be formatted before being input into the AI ​​model.

[1379] Analysis means

[1380] The server uses a generative AI model based on the preprocessed data to analyze the infant's needs. For example, if the facial and acoustic features match the pattern of "hungry," the AI ​​model will determine that the infant is "hungry." This analysis is performed using deep learning frameworks such as TensorFlow and PyTorch.

[1381] Notification means

[1382] The analysis results are converted into natural language on the server. For example, if the AI ​​model determines that the baby is hungry, it generates a message saying, "The baby is hungry. Please give him some milk." The generated message is sent to the device (smartphone) and notified to the user in real time. The notification is displayed on the smartphone screen and uses a notification sound, and is sent via Firebase Cloud Messaging or Apple Push Notification Service.

[1383] Emotion Engine

[1384] An emotion engine is a device or software that analyzes a user's facial expressions and behavior to recognize their emotional state. It uses emotion recognition libraries such as TensorFlow and OpenFace. For example, it identifies the user's stress level and emotional state (e.g., joy, anger, sadness). The emotion engine identifies the user's emotional state, and the notification method adjusts the message to take this into account. For example, if the user is stressed, an additional message such as "Please remain calm and respond calmly" is included.

[1385] Learning tools

[1386] The user acts according to the notification and records the results in the app. For example, the user can record the baby's reaction after feeding the baby. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the generative AI model, improving its analytical accuracy.

[1387] Specific examples

[1388] The user launches a smartphone app and points the camera at the baby's face. The smartphone or baby monitor begins recording video and audio in real time, and these data are sent to a server. The server extracts facial features from the video data and acoustic features from the audio data. A generative AI model analyzes these features and identifies the baby's request as "hungry." As a result, the server generates a message saying, "The baby is hungry. Please feed him / her." and sends it to the device. The device notifies the smartphone, and the user prepares the milk. The user then records the result in the app, and the device sends feedback data to the server. The server uses this data to retrain the generative AI model. Furthermore, an emotion engine analyzes the user's emotional state. If it determines that the user is stressed, it sends a message containing psychological support, such as, "The baby is hungry. Please feed him / her. Please stay calm and respond calmly."

[1389] Prompt Sentence Examples

[1390] "Please explain the data you would like to input into a generative AI model for a system that analyzes a baby's facial expressions and vocalizations to determine its needs and notify the parents."

[1391] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1392] Step 1:

[1393] The user launches the smartphone app and points the camera at the baby. When the user starts recording with the app, the smartphone's camera and microphone begin collecting video and audio of the baby. The input at this stage is the baby's real-time facial expressions and cries, and the output is video and audio data stored on the smartphone as digital data.

[1394] Step 2:

[1395] The smartphone (device) sends the collected video and audio data to a server using a communication method, specifically Wi-Fi or mobile data communication. The input at this stage is the video and audio data stored on the smartphone, and the output is digital data transferred to the server.

[1396] Step 3:

[1397] The server divides the received video data into frames and extracts the infant's facial position and facial expression features using a face detection algorithm (e.g., OpenCV or Dlib). It also extracts acoustic features (e.g., frequency spectrum and sound intensity) from the audio data using an audio processing library such as Librosa. The input at this stage is the video and audio data sent to the server, and the output is the processed facial and acoustic features.

[1398] Step 4:

[1399] The server inputs the preprocessed facial and acoustic features into a generative AI model to analyze the infant's needs. Specifically, the model performs analysis using deep learning frameworks such as TensorFlow and PyTorch to identify the infant's needs (e.g., "I'm hungry"). The input at this stage is the preprocessed feature data, and the output is the analyzed infant's needs.

[1400] Step 5:

[1401] The server converts the analysis results into natural language. This is achieved using an NLP engine such as GPT-3 or BERT. For example, if the analysis result is "hungry," the server generates a message saying "The baby is hungry. Please give him milk." The input at this stage is the analysis result as the infant's request, and the output is a notification message in natural language.

[1402] Step 6:

[1403] The server then sends the generated notification message to the device (smartphone). This is done via Firebase Cloud Messaging (FCM) or Apple Push Notification Service (APNs). The user receives the message in real time on their smartphone and is notified by a notification sound or a display on the screen. The input at this stage is a natural language notification message, and the output is a notification displayed on the smartphone.

[1404] Step 7:

[1405] The server records the user's facial expressions and behavior and analyzes their emotional state using an emotion engine (such as TensorFlow or OpenFace). If the server determines that the user is feeling stressed, it includes an additional psychological support message in the notification. The input of this stage is the user's facial expressions and voice data, and the output is a tailored notification message.

[1406] Step 8:

[1407] The user acts according to the notification and records the results in the app. For example, the user can enter the baby's reaction after feeding into the app. The device sends this feedback data to the server, which stores it in a database. The stored data is used to retrain the generative AI model. The input at this stage is the feedback data recorded by the user in the app, and the output is the retraining data stored on the server.

[1408] (Application example 2)

[1409] 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."

[1410] A system is needed that can not only accurately grasp the needs and status of infants and notify parents in real time, but also recognize the parents' emotional state and provide appropriate notifications according to the situation. In particular, there is a need for a system that can respond quickly to infants' needs while taking into consideration the parents' stress levels and emotional state, thereby reducing the burden of childcare and providing higher quality childcare support.

[1411] 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.

[1412] In this invention, the server includes data collection means for collecting the facial expressions and cries of the infant, data preprocessing means for preprocessing the collected facial expression and cries data, analysis means for analyzing the preprocessed data to identify the infant's needs, notification means for converting the analysis results into natural language and notifying the guardian, learning means for recording the infant's response results and relearning based on those results, and an emotion engine for recognizing the user's emotional state and adjusting the content of the notification. This makes it possible to accurately grasp the infant's needs and provide appropriate support according to the guardian's emotional state.

[1413] The "data collection means" is a device that includes a camera and a microphone to collect the infant's facial expressions and sounds.

[1414] The "data preprocessing means" is a means for executing a process to prepare the collected facial expression and vocalization data in a form suitable for analysis.

[1415] "Analysis means" means a means of using a generative AI model to identify the needs of an infant based on pre-processed data.

[1416] The "notification means" is a means for converting the analysis results into natural language and notifying the parents.

[1417] The "learning means" is a means for recording the infant's response results and using that information to improve the system's analytical accuracy.

[1418] The "emotion engine" is a means of recognizing the parent's emotional state and adjusting the content of notifications based on that information.

[1419] A "generative AI model" is an artificial intelligence model that identifies the needs and conditions of infants and young children based on collected and pre-processed data.

[1420] The system for implementing this invention collects and analyzes the facial expressions and cries of infants, and then notifies parents appropriately, thereby reducing the burden of childcare. It also recognizes the parents' emotional state and adjusts the content of notifications based on that information. This system utilizes various hardware and software and is implemented as follows:

[1421] Hardware and software used

[1422] Hardware: smartphone, smart glasses, server, camera and microphone.

[1423] Software: Data collection application (for smartphone / smart glasses), data pre-processing algorithm (server), emotion recognition model (generative AI model), notification application (smartphone).

[1424] System Overview

[1425] Data collection methods

[1426] The system uses a device (such as a smartphone or smart glasses) to collect the infant's facial expressions and cries. This uses a camera and microphone, and the collected video and audio data is sent to a server via the internet.

[1427] Data preprocessing measures

[1428] The server splits the video data into frames and performs face detection using the OpenCV library. It also extracts acoustic features such as frequency spectrum and volume from the audio data. These preprocessing steps prepare the data in a format suitable for analysis.

[1429] Analysis means

[1430] Based on the pre-processed data, a generative AI model is used to identify the infant's needs, such as whether they are hungry or sleepy. This analysis also utilizes past training data, so accuracy improves over time.

[1431] Notification means

[1432] The analysis results are converted into natural language on the server and sent to the parent's device. Notifications are displayed on the smartphone screen or via voice notification. The emotion engine also recognizes the parent's emotional state and adjusts the content of the notification based on that information. For example, if the parent is feeling stressed, a message containing psychological support will be sent.

[1433] Learning tools

[1434] Parents act on the advice provided and record their results in the app. This feedback data is sent to the server and used to retrain the model, further improving its analysis accuracy.

[1435] Specific example explanation

[1436] 1. Data Collection

[1437] Parents launch the smartphone app and take a picture of their baby's face with the camera. Audio data is also collected, and this data is sent to a server in real time.

[1438] 2. Data Preprocessing

[1439] The server preprocesses the data that arrives, extracting facial features from video data and acoustic features from audio data.

[1440] 3. Analysis

[1441] A generative AI model is used to analyze an infant's needs, for example, to identify the need to "be hungry."

[1442] 4. Notification

[1443] A message is generated and sent to the parent's smartphone saying, "Your baby is hungry. Please feed him / her." If the parent's emotional state indicates stress, an additional message is included saying, "Please stay calm."

[1444] Prompt Sentence Examples

[1445] Prompt: "Create a system that detects delivery workers' fatigue and suggests breaks. Collect facial expression data from the delivery workers using a camera and analyze the fatigue indicators using a model."

[1446] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1447] Step 1: Data collection

[1448] The user (guardian) starts the smartphone app and takes a picture of the baby's face with the camera. At the same time, audio is collected with the microphone. The collected video data (facial expressions) and audio data (cries) are sent from the device to the server in real time. This makes the input data available on the server.

[1449] Step 2: Data Preprocessing

[1450] The server splits the received video data into frames using the OpenCV library and performs face detection, extracting facial features. At the same time, acoustic features such as frequency spectrum and volume are extracted from the audio data. Once preprocessing is complete, these feature data become available as input data for analysis.

[1451] Step 3: Data analysis

[1452] The server inputs the preprocessed facial and acoustic features into a generative AI model. The generative AI model analyzes this data and identifies the infant's needs. For example, the request "I'm hungry" may be identified. The analysis results are output as a natural language message indicating a specific response.

[1453] Step 4: Notification

[1454] The server then sends the generated natural language message to the parent's smartphone. Notifications are displayed on the screen or sounded. The emotion engine then analyzes the parent's emotional state, and if it detects stress, it adds psychological support to the message. For example, it may include a message saying, "Please stay calm."

[1455] Step 5: Gather feedback

[1456] The user (guardian) takes action according to the notification and then records the results in the smartphone app. For example, they input "the baby's reaction after giving milk." This feedback data is sent from the device to the server.

[1457] Step 6: Retrain the model

[1458] The server saves the collected feedback data in a database and uses it to retrain the generative AI model, thereby improving the accuracy of analysis. As the feedback data accumulated sequentially is used as training data for the model, the accuracy of the model improves over time.

[1459] Based on this processing step, the system can accurately grasp the needs of infants and provide appropriate notifications that take into account the emotional state of the caregiver, thereby reducing the burden of childcare on the caregiver.

[1460] 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.

[1461] 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.

[1462] 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.

[1463] 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.

[1464] 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.

[1465] 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.

[1466] 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).

[1467] 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.

[1468] 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."

[1469] 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.

[1470] 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).

[1471] 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.

[1472] 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.

[1473] 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.

[1474] 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.

[1475] 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.

[1476] 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.

[1477] 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.

[1478] 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.

[1479] 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.

[1480] 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.

[1481] The following is further disclosed regarding the above embodiment.

[1482] (Claim 1)

[1483] a data collection means for collecting the infant's facial expressions and cries;

[1484] a data preprocessing means for preprocessing the collected facial expression and cry data;

[1485] analysis means for analyzing the pre-processed data to identify the infant's needs;

[1486] a notification means for converting the analysis results into natural language and notifying the parent;

[1487] A learning method for recording the infant's response results and relearning based on the results;

[1488] A system including:

[1489] (Claim 2)

[1490] 2. The system according to claim 1, wherein the data collection means includes a camera and a microphone, and collects the facial expressions and cries of the infant in real time.

[1491] (Claim 3)

[1492] The system according to claim 1, wherein the analysis means identifies the infant's needs using a generative AI model based on the extracted facial and acoustic features.

[1493] "Example 1"

[1494] (Claim 1)

[1495] a data collection means for collecting the infant's facial expressions and cries;

[1496] a data preprocessing means for preprocessing the collected facial expression and call data and extracting facial positions and acoustic features;

[1497] an analytical means for identifying the infant's needs using a generative AI model based on pre-processed data;

[1498] A notification means for converting the analysis results into natural language and notifying the parents in real time;

[1499] A learning method that records the infant's response results and retrains the generative AI model based on those results;

[1500] A system including:

[1501] (Claim 2)

[1502] 2. The system according to claim 1, wherein the data collection means includes a camera and a microphone, and collects the facial expressions and cries of the infant in real time and acquires them as digital data.

[1503] (Claim 3)

[1504] The system of claim 1, wherein the analysis means classifies the infant's needs using a generative AI model based on the extracted facial features and acoustic features.

[1505] "Application Example 1"

[1506] (Claim 1)

[1507] a data collection means for collecting facial expressions and cries of infants;

[1508] a data preprocessing means for preprocessing the collected facial expression and cry data;

[1509] analysis means for analyzing the pre-processed data to identify the infant's needs;

[1510] a notification means for converting the analysis results into natural language and notifying the parent;

[1511] A learning method for recording the results of the infant's response and re-learning based on the results;

[1512] a control means for controlling the driving of the vehicle based on the notification result from the analysis means;

[1513] a display means for displaying the request content to the driver when a request is made by the infant while driving;

[1514] A system including:

[1515] (Claim 2)

[1516] 2. The system according to claim 1, wherein the data collection means includes a camera and a microphone, and collects the facial expressions and cries of the infant in real time.

[1517] (Claim 3)

[1518] The system according to claim 1, wherein the analysis means identifies the infant's needs using a generative AI model based on the extracted facial features and acoustic features.

[1519] "Example 2: Combining Emotion Engines"

[1520] (Claim 1)

[1521] a data collection means for collecting the infant's facial expressions and cries;

[1522] a data preprocessing means for preprocessing the collected facial expression and cry data;

[1523] analysis means for analyzing the pre-processed data to identify the infant's needs;

[1524] a notification means for converting the analysis results into natural language and notifying the parent;

[1525] A learning method for recording the infant's response results and relearning based on the results;

[1526] an emotion engine that recognizes the user's emotional state and adjusts the notification content;

[1527] A system including:

[1528] (Claim 2)

[1529] 2. The system according to claim 1, wherein the data collection means includes a camera and a microphone, and collects the facial expressions and cries of the infant in real time.

[1530] (Claim 3)

[1531] The system according to claim 1, wherein the analysis means identifies the infant's needs using a generative AI model based on the extracted facial and acoustic features.

[1532] "Application example 2 when combining emotion engines"

[1533] (Claim 1)

[1534] a data collection means for collecting facial expressions and cries of infants;

[1535] a data preprocessing means for preprocessing the collected facial expression and cry data;

[1536] analysis means for analyzing the pre-processed data to identify the infant's needs;

[1537] a notification means for converting the analysis results into natural language and notifying the guardian;

[1538] A learning method for recording the results of the infant's response and re-learning based on the results;

[1539] A system that includes an emotion engine that recognizes the user's emotional state and adjusts the content of notifications.

[1540] (Claim 2)

[1541] 2. The system according to claim 1, wherein the data collection means includes a camera and a microphone, and collects the facial expressions and cries of the infant in real time.

[1542] (Claim 3)

[1543] The system according to claim 1, wherein the analysis means identifies the infant's needs using a generative AI model based on the extracted facial features and acoustic features. [Explanation of symbols]

[1544] 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 data collection means for collecting the infant's facial expressions and cries; a data preprocessing means for preprocessing the collected facial expression and cry data; analysis means for analyzing the pre-processed data to identify the infant's needs; a notification means for converting the analysis results into natural language and notifying the parent; A learning method for recording the infant's response results and relearning based on the results; A system including:

2. 2. The system according to claim 1, wherein the data collection means includes a camera and a microphone, and collects the facial expressions and cries of the infant in real time.

3. 2. The system according to claim 1, wherein the analyzing means identifies the infant's needs using a generative AI model based on the extracted facial and acoustic features.

Images