Middle and primary school emotional response type education auxiliary method and system and medium

Abstract

The invention discloses a middle and primary school emotion response type education auxiliary method and system and a medium, and the method comprises the steps: obtaining the facial expression data and voice data of a user, and extracting the facial emotion features of the facial expression data and the language emotion features of the voice data; analyzing the facial expression data and the voice data through an emotion recognition model, and performing learning emotion monitoring; and generating personalized teaching suggestions based on an emotional response mechanism. The emotional state of the student is accurately identified, the emotional change of the student is captured in real time, the teaching strategy is adjusted, and the emotional requirement and learning rhythm of the student are met.

Description

Technical Field

[0001] The present invention relates to the field of smart teaching technology, and in particular to an emotionally responsive education auxiliary method, system and medium for primary and secondary schools. Background Art

[0002] In the traditional primary and secondary education system, teaching activities usually focus on the transfer of knowledge and the cultivation of skills, but often ignore the profound impact of students' emotional state on learning outcomes. Studies have shown that students' emotional state not only directly affects their classroom participation and learning efficiency, but is also closely related to mental health. However, most current educational tools lack effective means of monitoring students' emotional states, and are unable to adjust teaching content and strategies in real time based on students' emotional feedback. This results in teachers being unable to capture students' emotional changes in real time and provide personalized teaching support for teachers to adjust teaching strategies to meet students' emotional needs and learning rhythm, resulting in poor teaching results.

[0003] With the rapid development of affective computing technology and artificial intelligence, how to apply these technologies to the field of education to improve teaching quality and improve students' mental health has become a hot topic in educational technology research. Existing primary and secondary school teaching systems are difficult to accurately identify students' emotional states, resulting in teachers being unable to capture students' emotional changes in real time and provide personalized teaching support for teachers to adjust teaching strategies to meet students' emotional needs and learning rhythm, resulting in poor teaching results. Summary of the invention

[0004] The present invention provides an emotion-responsive education auxiliary method, system and medium for primary and secondary schools, which can accurately identify students' emotional states and capture students' emotional changes in real time to adjust teaching strategies and meet students' emotional needs and learning rhythm.

[0005] In order to solve the above technical problems, the first aspect of the present invention discloses an emotional responsive education auxiliary method for primary and secondary schools, the method comprising:

[0006] Acquire the user's facial expression data and voice data, and extract the facial emotion features of the facial expression data and the language emotion features of the voice data;

[0007] Analyze the facial expression data and voice data through an emotion recognition model to perform learning emotion monitoring;

[0008] Generate personalized teaching suggestions based on emotional responsiveness mechanism.

[0009] In some embodiments, comprising:

[0010] Concatenate the facial emotion feature and the voice emotion feature to generate a comprehensive emotion feature vector;

[0011] The comprehensive emotion feature vector is input into the Transformer module and the fully connected layer for processing, and the output of the fully connected layer is converted into the probability distribution of multiple emotion states through the softmax activation function;

[0012] The emotional state of the user is determined according to the probability distribution.

[0013] In some implementations, extracting language emotion features of speech data includes:

[0014] Preprocessing the voice data;

[0015] Calculate a spectrogram by short-time Fourier transform, map the spectrogram to a Mel spectrogram, and calculate a logarithmic Mel spectrogram;

[0016] The logarithmic Mel spectrogram is input into the sound extraction model based on YAMNet, and the emotion score vector is output.

[0017] In some implementations, extracting facial emotion features of the facial expression data includes:

[0018] Preprocessing the facial expression data;

[0019] The preprocessed facial expression data is input into the facial emotion feature extraction model to extract facial emotion features.

[0020] In some embodiments, the personalized teaching suggestions include prompt lights, teaching strategies, teaching content, and / or teaching difficulty adjustments;

[0021] Generate personalized teaching suggestions based on emotional responsiveness, including:

[0022] Select a corresponding prompt light color according to the emotional state; the prompt light color includes a first color, a second color, and a third color;

[0023] Generate teaching strategies according to the emotional state, the teaching strategies include: guiding relaxation strategy, attention diversion strategy, emotional expression strategy, and individual counseling strategy;

[0024] When the emotional state continues to be negative in a number of monitoring cycles, teaching content with a teaching difficulty corresponding to negative emotions is selected.

[0025] In some embodiments, it further comprises:

[0026] Obtaining user historical data, including historical emotional state and historical personalized teaching suggestions;

[0027] Generate time series data based on the historical data, analyze emotional trends and learning status, and generate teaching adjustment strategies and teaching reports.

[0028] In some implementations, the facial emotion feature extraction model is:

[0029] F(x)=softmax(W 2 *ReLU(W 1 *(onv(x)+b 1 )+b 2 )

[0030] Where x is the input facial image; Conv(x) is a convolution operation to extract the features of the facial image, and W 1 , b 1 are the weights and biases of the convolutional layers, ReLU is the activation function; W 2 , b 2 are the weights and biases of the fully connected layer, and softmax is the activation function.

[0031] In some embodiments, the activation function is

[0032]

[0033] Among them, x i is the original output value of the i-th emotional state category, For x i index of all emotion categories sum.

[0034] According to a second aspect of the present invention, an emotionally responsive education auxiliary system for primary and secondary schools is disclosed, comprising:

[0035] The emotion recognition module obtains the user's facial expression data and voice data, and extracts the facial emotion features of the facial expression data and the language emotion features of the voice data;

[0036] A learning emotion monitoring module analyzes the facial expression data and the voice data through an emotion recognition model to perform learning emotion monitoring;

[0037] Personalized teaching module generates personalized teaching suggestions based on an emotional responsive mechanism.

[0038] In a third aspect, a computer storage medium is disclosed, characterized in that a computer program is stored thereon, and when the computer program is executed by a processor, the steps of any one of the above-mentioned emotionally responsive education assistance methods for primary and secondary schools are implemented.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] The present invention provides an emotion-responsive education auxiliary method, system, and medium for primary and secondary schools. By acquiring facial expression data and voice data, facial emotion features and language emotion features are extracted, the emotional state of students is identified in real time, and the emotional changes of students are captured in real time, so as to adjust the teaching strategy to meet the emotional needs and learning rhythm of students and improve the teaching effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 A schematic diagram of the process provided by the present invention;

[0042] Figure 2 A schematic diagram of a process for extracting language emotion features of speech data provided by the present invention;

[0043] Figure 3 A schematic diagram of a process for extracting facial emotion features from facial expression data provided by the present invention;

[0044] Figure 4 This is a schematic flow chart of step S2 provided by the present invention. DETAILED DESCRIPTION

[0045] For better understanding and implementation, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0046] The terms "including" and "having" and any variations thereof in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or modules is not necessarily limited to those steps or modules explicitly listed, but may include other steps or modules not explicitly listed or inherent to these processes, methods, products or apparatus.

[0047] The embodiment of the present invention discloses an emotion-responsive education auxiliary method for primary and secondary schools, which accurately identifies students' emotional states and captures students' emotional changes in real time to adjust teaching strategies and meet students' emotional needs and learning rhythm.

[0048] Specifically, Figure 1 As shown, the method comprises the following steps:

[0049] Step S1, obtaining the user's facial expression data and voice data, and extracting the facial emotion features of the facial expression data and the language emotion features of the voice data;

[0050] In this application, users can be parents, students, or teachers. Different students can learn from each other, parents or teachers can tutor students, or a single student can learn on his own. There is no limitation in this application.

[0051] When teachers provide learning guidance to students, the students’ facial expression data and voice data are collected through preset devices. Voice data includes intonation, volume, and speaking speed, which can further accurately analyze the user’s emotional state.

[0052] Further, such as Figure 2 As shown, extracting facial emotion features of the facial expression data includes:

[0053] Preprocessing the facial expression data;

[0054] The preprocessed facial expression data is input into the facial emotion feature extraction model to extract facial emotion features.

[0055] The facial expression data and speech data collected by the camera are preprocessed, including noise reduction, image enhancement and standardization, to ensure the quality of the input data and improve the clarity of the speech features. The preprocessed facial images are input into a deep neural network, such as ResNet, to extract the facial emotion features of the facial expression data, including emotion-related facial muscle movements, eye and mouth shape changes, etc.

[0056] RestNet uses multiple residual blocks to learn the feature representation of facial expression data. Each residual block includes a convolutional layer, batch normalization, and ReLU activation function. Skip connections are used to solve the gradient vanishing problem in deep network training.

[0057] Specifically, the facial emotion feature extraction model is:

[0058] F(x)=softmax(W 2 *ReLU(W 1 *(onv(x)+b 1 )+b 2 )

[0059] Where x is the input facial image; Conv(x) is a convolution operation to extract the features of the facial image, and W 1 , b 1 are the weights and biases of the convolutional layers, ReLU is the activation function; W 2 , b 2 is the weight and bias of the fully connected layer, and softmax is the activation function

[0060] The activation function is

[0061]

[0062] Among them, x i is the original output value of the i-th emotional state category, For x i The index of all categories sum.

[0063] The raw scores output by the neural network are converted into probability form through the softmax function to explain the probability prediction of each emotion category.

[0064] And as Figure 3 As shown, the language emotion features of the speech data are extracted, including:

[0065] Preprocessing the voice data;

[0066] Calculate a spectrogram by short-time Fourier transform, map the spectrogram to a Mel spectrogram, and calculate a logarithmic Mel spectrogram;

[0067] The logarithmic Mel spectrogram is input into the sound extraction model based on YAMNet, and the emotion score vector is output.

[0068] The preprocessing is to resample the input speech data into a mono 16kHz waveform to ensure the consistency of the model input. The preprocessed speech data is calculated through the short-time Fourier transform (STFT) to calculate the spectrogram, which is mapped to the Mel scale to obtain the Mel spectrogram.

[0069] Compute the log-mel spectrogram:

[0070] log_mel_spectrogram=log(mel_spectrogram+0.00 1)

[0071] Among them, mel_spectrogrammel_spectrogram is the Mel spectrogram, and 0.001 is added to avoid taking the logarithm of zero.

[0072] The log mel spectrogram is input into the sound extraction model based on YAMNet, and the sentiment score vector is output. The score vector represents the score of 521 audio event categories that may exist in the audio. The output of the model can be expressed as: scores = YAMNet (log_mel_spectrogram) where scores is a tensor with a shape of (N, 521) and N is the batch size.

[0073] Since the YAMNet pre-trained model was originally trained to identify audio events in AudioSet. In some embodiments, emotion recognition can be performed directly through voice data, an additional classifier is trained, which takes the score vector output by YAMNet as input and outputs the probability distribution of emotion categories through the softmax function: probabilities = softmax (W / scores + b) where W and b are the trained weights and biases, and probabilities is the probability distribution of emotion categories. The category with the highest probability is selected as the final emotional state.

[0074] Step S2: analyzing the facial expression data and the voice data through an emotion recognition model to perform learning emotion monitoring;

[0075] like Figure 4 As shown, the method includes:

[0076] Step S21, concatenating the facial emotion feature and the language emotion feature to generate a comprehensive emotion feature vector;

[0077] Step S22, inputting the comprehensive emotion feature vector into the Transformer module and the fully connected layer for processing, and converting the output of the fully connected layer into the probability distribution of multiple emotion states through the softmax activation function;

[0078] Step S23: determining the user's emotional state according to the probability distribution.

[0079] The facial emotion features and the language emotion features are connected in series to form a comprehensive emotion feature vector. By integrating facial emotion features and language emotion features, the emotion recognition model can utilize the complementarity of facial expression data and voice data to obtain a more complete emotional expression.

[0080] F combined =[F face ; F voice ]

[0081] Among them, F combined is the comprehensive sentiment feature vector, F face F is the facial emotion feature; voice It is the emotional feature of language.

[0082] The fused comprehensive emotional feature vector is input into the Transformer model for processing. The Transformer model can capture the complex relationship between features through the self-attention mechanism. For example, the association between facial expressions and voice features helps to identify more subtle emotional changes. Under the action of multi-head attention and feed-forward neural network, the Transformer module will generate more representative emotional feature representations.

[0083] The self-attention mechanism is as follows:

[0084]

[0085] Among them, Q, K, and V are query, key, and value matrices respectively, and d k is the dimension of the key vector.

[0086] Among them, the output of multi-head self-attention is the concatenation of all head outputs:

[0087] MultiHead(Q,K,V)=Concat(head 1 , ..., head h )W o

[0088] The output of each head is is the learnable weight matrix.

[0089] Each Transformer module also sets up a feed-forward network, which contains two linear transformations:

[0090] FFN(x)=max(0,xW 1 +b 1 )W 2 +b 2

[0091] Among them, W1, b1, w2, b2 are the weights and biases of the network layer.

[0092] The output of each self-attention layer and feedforward network is normalized and connected to the fully connected layer through residual connection to further integrate and compress the comprehensive sentiment feature vector.

[0093] Z=W fc ·F transformer +b fc

[0094] W fc With bfc is the weight and bias of the fully connected layer, F transformer is the output of the Transformer module.

[0095] The output of the fully connected layer is used to map the emotional features to a probability distribution through the Softmax activation function. Each emotion category corresponds to a probability, indicating the possibility that the student belongs to that emotion category. Emotion categories can include happiness, anger, sadness, surprise, fear, disgust, alertness, warning or danger, etc. The classification of emotion categories and the corresponding probabilities can be adjusted according to the actual application scenario, which is not limited in this application.

[0096] According to the output probability of the Softmax activation function, the emotion category with the highest probability is selected as the user's current emotional state. The specific emotional state of the students in the learning process is identified, such as happiness, anger, sadness, surprise, fear, disgust or neutrality, and the emotional state is used for emotional feedback and management in the teaching process.

[0097] Step S3: Generate personalized teaching suggestions based on the emotion-responsive mechanism.

[0098] Generate personalized teaching suggestions based on emotional responsiveness, including:

[0099] Select a corresponding prompt light color according to the emotional state; the prompt light color includes a first color, a second color, and a third color;

[0100] Generate teaching strategies according to the emotional state, the teaching strategies include: guiding relaxation strategy, attention diversion strategy, emotional expression strategy, and individual counseling strategy;

[0101] When the emotional state continues to be negative in a number of monitoring cycles, teaching content with a teaching difficulty corresponding to negative emotions is selected.

[0102] For example, the correspondence between preset colors and students' emotions is indicated by the color of the indicator light. The colors include green, yellow, and red. Green represents that the student's emotions are stable; yellow means that the student may be distracted or confused and needs the teacher's attention; red means that the student may be in a negative emotion such as anger or disgust and needs to adjust his emotions immediately. The color of the indicator light provides real-time feedback on the student's emotional state, allowing teachers to understand the student's emotional changes in a timely manner and quickly adjust teaching strategies.

[0103] When the student's emotional status is shown as green, the teacher can continue the current teaching rhythm while encouraging the student to maintain a positive learning attitude.

[0104] When the affective status is shown as yellow, teachers can slow down their teaching, provide more explanations and examples, or invite students to ask questions to help them understand difficult points.

[0105] When the emotional status is shown as red, it is necessary to focus on the student's emotions and adjust the teaching strategy, including:

[0106] Guided relaxation strategies: Guide students in deep breathing exercises or brief meditation to help them relax.

[0107] Attention diversion strategy: Divert students' attention and relieve negative emotions by introducing an interesting experiment or story.

[0108] Emotional Expression Strategies: Provide a safe environment for students to express their feelings and for teachers to listen and provide support.

[0109] Positive feedback strategy: Give students positive feedback and encouragement to help them rebuild their confidence.

[0110] Individual counseling strategies: Conduct one-on-one counseling with students after class to understand their emotional problems and provide personalized support.

[0111] Furthermore, if a student's emotional state continues to be red for several monitoring cycles, the teacher can immediately adjust the teaching strategy, such as pausing the current course content and switching to a group discussion or interactive game to relieve the student's nervousness. At the same time, the teacher can communicate with the students privately to understand their feelings and provide appropriate emotional support and guidance.

[0112] For example, when a student's negative emotions (such as disgust, anger) are detected, the teaching content can be automatically adjusted to make it more in line with the student's interests and needs. For example, if a student shows frustration in a math class, more interesting math games or visual teaching materials can be recommended to increase student participation and interest.

[0113] Recommend different teaching methods based on students’ emotional states. For example, for students who show anxiety or tension, group discussions or cooperative learning can be used to reduce individual stress.

[0114] For students who show excitement or curiosity, inquiry-based learning or project-based learning can be introduced to encourage students to actively explore and solve problems.

[0115] The teaching difficulty can be adjusted dynamically according to students’ emotional feedback and learning behaviors. For example, for students who show confusion or frustration, the teaching difficulty can be reduced and more guidance and support can be provided.

[0116] For students who show confidence and interest, increase the teaching difficulty and provide more challenging tasks to stimulate students' potential.

[0117] Furthermore, the method further comprises:

[0118] Obtaining user historical data, including historical emotional state and historical personalized teaching suggestions;

[0119] Generate time series data based on the historical data, analyze emotional trends and learning status, and generate teaching adjustment strategies and teaching reports.

[0120] By recording students' historical emotional states, teachers can track students' emotional trends. Use statistical methods such as mean, median, standard deviation, etc. to analyze emotional data and determine the general trend of students' emotions. Apply time series analysis techniques (ARIMA model) to predict emotional trends and identify any abnormal or significant emotional fluctuations. Time series data can show the trend of user emotions and the impact of personalized teaching suggestions on students' emotions and learning outcomes.

[0121] Based on emotional trends, combined with learning achievement data such as historical test scores and homework submissions, students’ learning status is assessed, and cluster analysis is used to identify groups of students with similar emotions and learning performance. Based on students’ emotional status, appropriate teaching content and methods are automatically recommended. Based on students’ emotional status and learning behavior, appropriate teaching methods are automatically recommended, such as adjusting teaching difficulty, introducing interactive games, or using group discussions.

[0122] The teaching report includes emotional trend charts, learning status summaries and teaching strategy suggestions to help teachers adjust their teaching methods. It can help teachers and parents understand students' performance in different situations and provide data support for the formulation of effective teaching plans.

[0123] The present invention provides an emotion-responsive education auxiliary method for primary and secondary schools. By acquiring facial expression data and voice data, facial emotion features and language emotion features are extracted, the emotional state of students is identified in real time, and the emotional changes of students are captured in real time, so as to adjust the teaching strategy to meet the emotional needs and learning rhythm of students and improve the teaching effect.

[0124] Based on the same inventive concept, the present application provides an emotionally responsive educational assistance system for primary and secondary schools, including:

[0125] The emotion recognition module obtains the user's facial expression data and voice data, and extracts the facial emotion features of the facial expression data and the language emotion features of the voice data;

[0126] A learning emotion monitoring module analyzes the facial expression data and the voice data through an emotion recognition model to perform learning emotion monitoring;

[0127] Personalized teaching module generates personalized teaching suggestions based on an emotional responsive mechanism.

[0128] The processing method of this system can refer to the description of the above method and will not be repeated here.

[0129] The present invention provides a method, system and medium for emotionally responsive education assistance in primary and secondary schools, which can obtain the emotional state and behavior of users in real time and accurately identify the current emotional state and behavior of users. Compared with traditional methods, it helps to timely identify negative emotions and inappropriate behaviors during learning and tutoring, and helps parents maintain proper emotional management during tutoring their children.

[0130] The emotionally responsive education auxiliary methods, systems, and media provided in this application have market prospects and application value, and are suitable for: 1. Primary and secondary education institutions, to improve the quality of classroom teaching and the mental health level of students; 2. Online education platforms, to provide personalized online teaching support; 3. Family education scenarios, to help parents better guide their children's learning; 4. Higher education and other training fields, with broad application prospects.

[0131] The present invention also provides a device, which may include: a memory storing executable program code;

[0132] a processor coupled to the memory;

[0133] A transceiver used to communicate with other devices or communication networks and to receive or send network messages;

[0134] A bus used to connect memory, processors, and transceivers for internal communication.

[0135] The transceiver receives messages transmitted over the network and passes them to the processor via the bus. The processor calls the executable program code stored in the memory through the bus for processing, and passes the processing results to the transceiver via the bus for sending, thereby realizing the method provided in the embodiment of the present application.

[0136] An embodiment of the present application also provides a non-transitory machine-readable storage medium, on which an executable program is stored. When the executable program is executed by a processor, the processor executes the method provided in the above embodiment.

[0137] An embodiment of the present invention discloses a computer-readable storage medium storing a computer program for electronic data exchange, wherein the computer program enables a computer to execute the described method.

[0138] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to enable a computer to execute the described method.

[0139] The embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without paying creative labor.

[0140] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0141] Finally, it should be noted that the embodiments disclosed in the present invention are only preferred embodiments of the present invention, which are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A method for assisting primary and secondary school emotional responsive education, characterized in that: The method comprises: Acquire the user's facial expression data and voice data, and extract the facial emotion features of the facial expression data and the language emotion features of the voice data; Analyze the facial expression data and voice data through an emotion recognition model to perform learning emotion monitoring; Generate personalized teaching suggestions based on emotional responsiveness mechanism.

2. The method for emotionally responsive education in primary and secondary schools according to claim 1, characterized in that: The facial expression data and the voice data are analyzed by an emotion recognition model to perform learning emotion monitoring, including: Concatenate the facial emotion features and the language emotion features to generate a comprehensive emotion feature vector; The comprehensive emotional feature vector is input into the Transformer module and the fully connected layer for processing, and the output of the fully connected layer is converted into the probability distribution of multiple emotional states through the softmax activation function; The emotional state of the user is determined according to the probability distribution.

3. The method for emotionally responsive education in primary and secondary schools according to claim 1, characterized in that: Extract the language emotion features of speech data, including: Preprocessing the voice data; Calculate a spectrogram by short-time Fourier transform, map the spectrogram to a Mel spectrogram, and calculate a logarithmic Mel spectrogram; The logarithmic Mel spectrogram is input into the sound extraction model based on YAMNet, and the emotion score vector is output.

4. The method for assisting primary and secondary school emotional responsive education according to claim 1, characterized in that: Extracting facial emotion features of the facial expression data includes: Preprocessing the facial expression data; The preprocessed facial expression data is input into the facial emotion feature extraction model to extract facial emotion features.

5. The method for emotionally responsive education in primary and secondary schools according to claim 2, characterized in that: The personalized teaching suggestions include prompt lights, teaching strategies, teaching content and / or teaching difficulty adjustment; Generate personalized teaching suggestions based on emotional responsiveness, including: Select a corresponding prompt light color according to the emotional state; the prompt light color includes a first color, a second color, and a third color; Generate teaching strategies according to the emotional state, the teaching strategies include: guiding relaxation strategy, attention diversion strategy, emotional expression strategy, and individual counseling strategy; When the emotional state continues to be negative in a number of monitoring cycles, teaching content with a teaching difficulty corresponding to negative emotions is selected.

6. The method for emotionally responsive education in primary and secondary schools according to claim 1, characterized in that: Also includes: Obtaining user historical data, including historical emotional state and historical personalized teaching suggestions; Generate time series data based on the historical data, analyze emotional trends and learning status, and generate teaching adjustment strategies and teaching reports.

7. The method for emotionally responsive education in primary and secondary schools according to claim 4, characterized in that: The facial emotion feature extraction model is: F(x)=softmax(W2*ReLU(W1*Conv(x)+b1)+b2) Wherein, x is the input facial image; Conv(x) is a convolution operation to extract the features of the facial image, W1, b1 are the weights and biases of the convolutional layers respectively, and ReLU is the activation function; W2, b2 are the weights and biases of the fully connected layer, and softmax is the activation function.

8. The method for emotionally responsive education in primary and secondary schools according to claim 7, characterized in that: The activation function is Among them, x i is the original output value of the i-th emotional state category, For x i index of all emotion categories sum.

9. An emotionally responsive education assistance system for primary and secondary schools, characterized in that: include: The emotion recognition module obtains the user's facial expression data and voice data, and extracts the facial emotion features of the facial expression data and the language emotion features of the voice data; A learning emotion monitoring module analyzes the facial expression data and the voice data through an emotion recognition model to perform learning emotion monitoring; Personalized teaching module generates personalized teaching suggestions based on an emotional responsive mechanism.

10. A computer storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the steps of an emotionally responsive education assistance method for primary and secondary schools as claimed in any one of claims 1-8 are implemented.

Images