Facial Expression Recognition with High Response-Based Local Directional Pattern (HR-LDP) Network

Although lots of research has been done in recognizing facial expressions,there is still a need to increase the accuracy of facial expression recognition,particularly under uncontrolled situations.The use of Local Directional Patterns(LDP),which has good characteristics for emotion detection has yielded encouraging results.An innova-tive end-to-end learnable High Response-based Local Directional Pattern(HR-LDP)network for facial emotion recognition is implemented by employing fixed convolutional filters in the proposed work.By combining learnable convolutional layers with fixed-parameter HR-LDP layers made up of eight Kirsch filters and derivable simulated gate functions,this network considerably minimizes the number of network parameters.The cost of the parameters in our fully linked layers is up to 64 times lesser than those in currently used deep learning-based detection algorithms.On seven well-known databases,including JAFFE,CK+,MMI,SFEW,OULU-CASIA and MUG,the recognition rates for seven-class facial expression recognition are 99.36%,99.2%,97.8%,60.4%,91.1%and 90.1%,respectively.The results demonstrate the advantage of the proposed work over cutting-edge techniques.

Faster Region Convolutional Neural Network(FRCNN)Based Facial Emotion Recognition

Facial emotion recognition(FER)has become a focal point of research due to its widespread applications,ranging from human-computer interaction to affective computing.While traditional FER techniques have relied on handcrafted features and classification models trained on image or video datasets,recent strides in artificial intelligence and deep learning(DL)have ushered in more sophisticated approaches.The research aims to develop a FER system using a Faster Region Convolutional Neural Network(FRCNN)and design a specialized FRCNN architecture tailored for facial emotion recognition,leveraging its ability to capture spatial hierarchies within localized regions of facial features.The proposed work enhances the accuracy and efficiency of facial emotion recognition.The proposed work comprises twomajor key components:Inception V3-based feature extraction and FRCNN-based emotion categorization.Extensive experimentation on Kaggle datasets validates the effectiveness of the proposed strategy,showcasing the FRCNN approach’s resilience and accuracy in identifying and categorizing facial expressions.The model’s overall performance metrics are compelling,with an accuracy of 98.4%,precision of 97.2%,and recall of 96.31%.This work introduces a perceptive deep learning-based FER method,contributing to the evolving landscape of emotion recognition technologies.The high accuracy and resilience demonstrated by the FRCNN approach underscore its potential for real-world applications.This research advances the field of FER and presents a compelling case for the practicality and efficacy of deep learning models in automating the understanding of facial emotions.

Multi-Objective Equilibrium Optimizer for Feature Selection in High-Dimensional English Speech Emotion Recognition

Speech emotion recognition(SER)uses acoustic analysis to find features for emotion recognition and examines variations in voice that are caused by emotions.The number of features acquired with acoustic analysis is extremely high,so we introduce a hybrid filter-wrapper feature selection algorithm based on an improved equilibrium optimizer for constructing an emotion recognition system.The proposed algorithm implements multi-objective emotion recognition with the minimum number of selected features and maximum accuracy.First,we use the information gain and Fisher Score to sort the features extracted from signals.Then,we employ a multi-objective ranking method to evaluate these features and assign different importance to them.Features with high rankings have a large probability of being selected.Finally,we propose a repair strategy to address the problem of duplicate solutions in multi-objective feature selection,which can improve the diversity of solutions and avoid falling into local traps.Using random forest and K-nearest neighbor classifiers,four English speech emotion datasets are employed to test the proposed algorithm(MBEO)as well as other multi-objective emotion identification techniques.The results illustrate that it performs well in inverted generational distance,hypervolume,Pareto solutions,and execution time,and MBEO is appropriate for high-dimensional English SER.

How to enhance agricultural plastic waste management in China?Insights from public participation

Agricultural plastics play a pivotal role in agricultural production.However,due to expensive costs,agricultural plastic waste management(APWM)encounters a vast funding gap.As one of the crucial stakeholders,the public deserves to make appropriate efforts for APWM.Accordingly,identifying whether the public is willing to pay for APWM and clarifying the decisions’driving pathways to explore initiatives for promoting their payment intentions are essential to address the dilemma confronting APWM.To this end,by applying the extended theory of planned behavior(TPB),the study conducted an empirical analysis based on 1,288 residents from four provinces(autonomous regions)of northern China.Results illustrate that:1)respondents hold generally positive and relatively strong payment willingness towards APWM;2)respondents’attitude(AT),subjective norm(SN),and perceived behavioral control(PBC)are positively correlated with their payment intentions(INT);3)environmental cognition(EC)and environmental emotion(EE)positively moderate the relationships between AT and INT,and between SN and INT,posing significant indirect impacts on INT.The study’s implications extend to informing government policies,suggesting that multi-entity cooperation,specifically public payment for APWM,can enhance agricultural non-point waste management.

E2E-MFERC:AMulti-Face Expression Recognition Model for Group Emotion Assessment

In smart classrooms, conducting multi-face expression recognition based on existing hardware devices to assessstudents’ group emotions can provide educators with more comprehensive and intuitive classroom effect analysis,thereby continuouslypromotingthe improvementof teaching quality.However,most existingmulti-face expressionrecognition methods adopt a multi-stage approach, with an overall complex process, poor real-time performance,and insufficient generalization ability. In addition, the existing facial expression datasets are mostly single faceimages, which are of low quality and lack specificity, also restricting the development of this research. This paperaims to propose an end-to-end high-performance multi-face expression recognition algorithm model suitable forsmart classrooms, construct a high-quality multi-face expression dataset to support algorithm research, and applythe model to group emotion assessment to expand its application value. To this end, we propose an end-to-endmulti-face expression recognition algorithm model for smart classrooms (E2E-MFERC). In order to provide highqualityand highly targeted data support for model research, we constructed a multi-face expression dataset inreal classrooms (MFED), containing 2,385 images and a total of 18,712 expression labels, collected from smartclassrooms. In constructing E2E-MFERC, by introducing Re-parameterization visual geometry group (RepVGG)block and symmetric positive definite convolution (SPD-Conv) modules to enhance representational capability;combined with the cross stage partial network fusion module optimized by attention mechanism (C2f_Attention),it strengthens the ability to extract key information;adopts asymptotic feature pyramid network (AFPN) featurefusion tailored to classroomscenes and optimizes the head prediction output size;achieves high-performance endto-end multi-face expression detection. Finally, we apply the model to smart classroom group emotion assessmentand provide design references for classroom effect analysis evaluation metrics. Experiments based on MFED showthat the mAP and F1-score of E2E-MFERC on classroom evaluation data reach 83.6% and 0.77, respectively,improving the mAP of same-scale You Only Look Once version 5 (YOLOv5) and You Only Look Once version8 (YOLOv8) by 6.8% and 2.5%, respectively, and the F1-score by 0.06 and 0.04, respectively. E2E-MFERC modelhas obvious advantages in both detection speed and accuracy, which can meet the practical needs of real-timemulti-face expression analysis in classrooms, and serve the application of teaching effect assessment very well.

Exploring Sequential Feature Selection in Deep Bi-LSTM Models for Speech Emotion Recognition

Machine Learning(ML)algorithms play a pivotal role in Speech Emotion Recognition(SER),although they encounter a formidable obstacle in accurately discerning a speaker’s emotional state.The examination of the emotional states of speakers holds significant importance in a range of real-time applications,including but not limited to virtual reality,human-robot interaction,emergency centers,and human behavior assessment.Accurately identifying emotions in the SER process relies on extracting relevant information from audio inputs.Previous studies on SER have predominantly utilized short-time characteristics such as Mel Frequency Cepstral Coefficients(MFCCs)due to their ability to capture the periodic nature of audio signals effectively.Although these traits may improve their ability to perceive and interpret emotional depictions appropriately,MFCCS has some limitations.So this study aims to tackle the aforementioned issue by systematically picking multiple audio cues,enhancing the classifier model’s efficacy in accurately discerning human emotions.The utilized dataset is taken from the EMO-DB database,preprocessing input speech is done using a 2D Convolution Neural Network(CNN)involves applying convolutional operations to spectrograms as they afford a visual representation of the way the audio signal frequency content changes over time.The next step is the spectrogram data normalization which is crucial for Neural Network(NN)training as it aids in faster convergence.Then the five auditory features MFCCs,Chroma,Mel-Spectrogram,Contrast,and Tonnetz are extracted from the spectrogram sequentially.The attitude of feature selection is to retain only dominant features by excluding the irrelevant ones.In this paper,the Sequential Forward Selection(SFS)and Sequential Backward Selection(SBS)techniques were employed for multiple audio cues features selection.Finally,the feature sets composed from the hybrid feature extraction methods are fed into the deep Bidirectional Long Short Term Memory(Bi-LSTM)network to discern emotions.Since the deep Bi-LSTM can hierarchically learn complex features and increases model capacity by achieving more robust temporal modeling,it is more effective than a shallow Bi-LSTM in capturing the intricate tones of emotional content existent in speech signals.The effectiveness and resilience of the proposed SER model were evaluated by experiments,comparing it to state-of-the-art SER techniques.The results indicated that the model achieved accuracy rates of 90.92%,93%,and 92%over the Ryerson Audio-Visual Database of Emotional Speech and Song(RAVDESS),Berlin Database of Emotional Speech(EMO-DB),and The Interactive Emotional Dyadic Motion Capture(IEMOCAP)datasets,respectively.These findings signify a prominent enhancement in the ability to emotional depictions identification in speech,showcasing the potential of the proposed model in advancing the SER field.

Multimodal Sentiment Analysis Based on a Cross-Modal Multihead Attention Mechanism

Multimodal sentiment analysis aims to understand people’s emotions and opinions from diverse data.Concate-nating or multiplying various modalities is a traditional multi-modal sentiment analysis fusion method.This fusion method does not utilize the correlation information between modalities.To solve this problem,this paper proposes amodel based on amulti-head attention mechanism.First,after preprocessing the original data.Then,the feature representation is converted into a sequence of word vectors and positional encoding is introduced to better understand the semantic and sequential information in the input sequence.Next,the input coding sequence is fed into the transformer model for further processing and learning.At the transformer layer,a cross-modal attention consisting of a pair of multi-head attention modules is employed to reflect the correlation between modalities.Finally,the processed results are input into the feedforward neural network to obtain the emotional output through the classification layer.Through the above processing flow,the model can capture semantic information and contextual relationships and achieve good results in various natural language processing tasks.Our model was tested on the CMU Multimodal Opinion Sentiment and Emotion Intensity(CMU-MOSEI)and Multimodal EmotionLines Dataset(MELD),achieving an accuracy of 82.04% and F1 parameters reached 80.59% on the former dataset.

Impact of propofol and sevoflurane anesthesia on cognition and emotion in gastric cancer patients undergoing radical resection

BACKGROUND Propofol and sevoflurane are commonly used anesthetic agents for maintenance anesthesia during radical resection of gastric cancer.However,there is a debate concerning their differential effects on cognitive function,anxiety,and depression in patients undergoing this procedure.AIM To compare the effects of propofol and sevoflurane anesthesia on postoperative cognitive function,anxiety,depression,and organ function in patients undergoing radical resection of gastric cancer.METHODS A total of 80 patients were involved in this research.The subjects were divided into two groups:Propofol group and sevoflurane group.The evaluation scale for cognitive function was the Loewenstein occupational therapy cognitive assessment(LOTCA),and anxiety and depression were assessed with the aid of the self-rating anxiety scale(SAS)and self-rating depression scale(SDS).Hemodynamic indicators,oxidative stress levels,and pulmonary function were also measured.RESULTS The LOTCA score at 1 d after surgery was significantly lower in the propofol group than in the sevoflurane group.Additionally,the SAS and SDS scores of the sevoflurane group were significantly lower than those of the propofol group.The sevoflurane group showed greater stability in heart rate as well as the mean arterial pressure compared to the propofol group.Moreover,the sevoflurane group displayed better pulmonary function and less lung injury than the propofol group.CONCLUSION Both propofol and sevoflurane could be utilized as maintenance anesthesia during radical resection of gastric cancer.Propofol anesthesia has a minimal effect on patients'pulmonary function,consequently enhancing their postoperative recovery.Sevoflurane anesthesia causes less impairment on patients'cognitive function and mitigates negative emotions,leading to an improved postoperative mental state.Therefore,the selection of anesthetic agents should be based on the individual patient's specific circumstances.

Effect of a systemic intervention combined with a psychological intervention in stroke patients with oropharyngeal dysfunction

BACKGROUND Stroke frequently results in oropharyngeal dysfunction(OD),leading to difficulties in swallowing and eating,as well as triggering negative emotions,malnutrition,and aspiration pneumonia,which can be detrimental to patients.However,routine nursing interventions often fail to address these issues adequately.Systemic and psychological interventions can improve dysphagia symptoms,relieve negative emotions,and improve quality of life.However,there are few clinical reports of systemic interventions combined with psychological interventions for stroke patients with OD.AIM To explore the effects of combining systemic and psychological interventions in stroke patients with OD.METHODS This retrospective study included 90 stroke patients with OD,admitted to the Second Affiliated Hospital of Qiqihar Medical College(January 2022–December 2023),who were divided into two groups:regular and coalition.Swallowing function grading(using a water swallow test),swallowing function[using the standardized swallowing assessment(SSA)],negative emotions[using the selfrating anxiety scale(SAS)and self-rating depression scale(SDS)],and quality of life(SWAL-QOL)were compared between groups before and after the intervention;aspiration pneumonia incidence was recorded.RESULTS Post-intervention,the coalition group had a greater number of patients with grade 1 swallowing function compared to the regular group,while the number of patients with grade 5 swallowing function was lower than that in the regular group(P<0.05).Post-intervention,the SSA,SAS,and SDS scores of both groups decreased,with a more significant decrease observed in the coalition group(P<0.05).Additionally,the total SWAL-QOL score in both groups increased,with a more significant increase observed in the coalition group(P<0.05).During the intervention period,the total incidence of aspiration and aspiration pneumonia in the coalition group was lower than that in the control group(4.44%vs 20.00%;P<0.05).CONCLUSION Systemic intervention combined with psychological intervention can improve dysphagia symptoms,alleviate negative emotions,enhance quality of life,and reduce the incidence of aspiration pneumonia in patients with OD.

Impaired implicit emotion regulation in patients with panic disorder:An event-related potential study on affect labeling

BACKGROUND Panic disorder(PD)involves emotion dysregulation,but its underlying mechanisms remain poorly understood.Previous research suggests that implicit emotion regulation may play a central role in PD-related emotion dysregulation and symptom maintenance.However,there is a lack of studies exploring the neural mechanisms of implicit emotion regulation in PD using neurophysiological indicators.AIM To study the neural mechanisms of implicit emotion regulation in PD with eventrelated potentials(ERP).METHODS A total of 25 PD patients and 20 healthy controls(HC)underwent clinical evaluations.The study utilized a case-control design with random sampling,selecting participants for the case group from March to December 2018.Participants performed an affect labeling task,using affect labeling as the experimental condition and gender labeling as the control condition.ERP and behavioral data were recorded to compare the late positive potential(LPP)within and between the groups.RESULTS Both PD and HC groups showed longer reaction times and decreased accuracy under the affect labeling.In the HC group,late LPP amplitudes exhibited a dynamic pattern of initial increase followed by decrease.Importantly,a significant group×condition interaction effect was observed.Simple effect analysis revealed a reduction in the differences of late LPP amplitudes between the affect labeling and gender labeling conditions in the PD group compared to the HC group.Furthermore,among PD patients under the affect labeling,the late LPP was negatively correlated with disease severity,symptom frequency,and intensity.CONCLUSION PD patients demonstrate abnormalities in implicit emotion regulation,hampering their ability to mobilize cognitive resources for downregulating negative emotions.The late LPP amplitude in response to affect labeling may serve as a potentially valuable clinical indicator of PD severity.

Emotion Measurement Using Biometric Signal

In recent years, research on the estimation of human emotions has been active, and its application is expected in various fields. Biological reactions, such as electroencephalography (EEG) and root mean square successive difference (RMSSD), are indicators that are less influenced by individual arbitrariness. The present study used EEG and RMSSD signals to assess the emotions aroused by emotion-stimulating images in order to investigate whether various emotions are associated with characteristic biometric signal fluctuations. The participants underwent EEG and RMSSD while viewing emotionally stimulating images and answering the questionnaires. The emotions aroused by emotionally stimulating images were assessed by measuring the EEG signals and RMSSD values to determine whether different emotions are associated with characteristic biometric signal variations. Real-time emotion analysis software was used to identify the evoked emotions by describing them in the Circumplex Model of Affect based on the EEG signals and RMSSD values. Emotions other than happiness did not follow the Circumplex Model of Affect in this study. However, ventral attentional activity may have increased the RMSSD value for disgust as the β/θ value increased in right-sided brain waves. Therefore, the right-sided brain wave results are necessary when measuring disgust. Happiness can be assessed easily using the Circumplex Model of Affect for positive scene analysis. Improving the current analysis methods may facilitate the investigation of face-to-face communication in the future using biometric signals.

Enhancing Women's Well-being and Self-Confidence through Fragrance Cosmetics:Perceptions within Close and Broader Social Circles

Mental and physical stress can affect people's self-confidence and happiness.The primary objective of this study is to develop an active fragrance that has been proved to have a positive effect on happiness and self-confidence based on the latest research results of aroma components and fragrance innovation.The active fragrance is applied to the cream formula to study the emotions it brings to the user and extend the research scope of the active emotion to the close and extensive social circle of users to see whether it can perceive it.This study provides a reference cosmetic solution for emotional regulating active fragrance.

Film and Television Website Scores Authenticity Verification Based on the Emotional Analysis

Sentiment analysis is a method to identify and understand the emotion in the text through NLP and text analysis. In the era of information technology, there is often a certain error between the comments on the movie website and the actual score of the movie, and sentiment analysis technology provides a new way to solve this problem. In this paper, Python is used to obtain the movie review data from the Douban platform, and the model is constructed and trained by using naive Bayes and Bi-LSTM. According to the index, a better Bi-LSTM model is selected to classify the emotion of users’ movie reviews, and the classification results are scored according to the classification results, and compared with the real ratings on the website. According to the error of the final comparison results, the feasibility of this technology in the scoring direction of film reviews is being verified. By applying this technology, the phenomenon of film rating distortion in the information age can be prevented and the rights and interests of film and television works can be safeguarded.

Analysis of the Effect of Outpatient Nursing Intervention on Hypoglycemic Treatment Effect and Psychological Emotions of Diabetic Patients

Objective:To explore the effect of outpatient nursing interventions on the hypoglycemic treatment and psychological emotions of diabetic patients.Methods:148 patients who came to our hospital for outpatient treatment from February 2022 to October 2023 were selected and divided into a control group and an observation group,with 74 cases per group,according to the random number table method.The control group received routine nursing intervention,and the observation group received outpatient nursing intervention based on the control group.The two groups were observed for their effects of hypoglycemic treatment and psychological and emotional improvement before and after outpatient nursing intervention.Results:The health behavior scores of the control group were lower than that of the observation group;the post-intervention fasting blood glucose,2h postprandial blood glucose,anxiety self-rating scale(SAS),and the depression self-rating scale(SDS)of the control group were significantly higher than that of the observation group,and the difference was statistically significant(P<0.01).Conclusion:Outpatient nursing intervention encouraged patients to comply with healthy behaviors and helped control blood sugar levels.Patients’anxiety,depression,and other adverse psychological states were also improved hence the outpatient nursing intervention is worthy of further promotion.

Analysis of the Clinical Effect of Carisolv Minimally Invasive Gel in the Treatment of Pediatric Dental Caries

Objective:To study the clinical effect of Carisolv minimally invasive gel in the treatment of pediatric dental caries and its effect on pain.Methods:The research subjects of this paper were 113 cases of pediatric caries admitted to the hospital from April 2021 to April 2023,which were divided into two groups by the randomized table method.The control group(n=56)received the traditional dental drilling treatment method,and the observation group(n=57)applied Carisolv minimally invasive gel for treatment.The pain sensitivity and clinical efficacy as well as the emotions and adherence of the children were compared between the two groups.Results:The emotional score(ES)of children in the observation group was significantly lower than that of the control group,and the Frankl Adherence Scale score was significantly higher than that of the control group,P<0.05;the pain sensitivity of children in the observation group was better than that of the control group,and the total clinical efficacy rate of children in the observation group was significantly higher than that of the control group,P<0.05.Conclusion:Carisolv minimally invasive gel has considerable efficacy in the treatment of pediatric caries,and it can alleviate pain and improve children’s emotional state and adherence to the program.Thus,it is suitable for wide clinical applications.

A Review of Research on Second Language Acquisition from a Positive Psychology Perspective

This paper reviews the research on second language acquisition from the perspective of positive psychology.First,it introduces the background and purpose of the study and discusses the significance of the application of positive psychology in the field of language acquisition.Then,the basic theories of positive psychology,including the core concepts and principles of positive psychology,are summarized.Subsequently,the theory of second language acquisition is defined and outlined,including the definition,characteristics,and related developmental theories of second language acquisition.On this basis,the study of second language acquisition from the perspective of positive psychology is discussed in detail.By combing and synthesizing the literature,this paper summarizes the current situation and trend of second language acquisition research under the perspective of positive psychology and puts forward some future research directions and suggestions.

Pedestrian evacuation simulation in multi-exit case:An emotion and group dual-driven method

This paper analyzes the characteristics of emotion state and group behavior in the evacuation process.During the emergency evacuation,emotion state and group behavior are interacting with each other,and indivisible.The emotion spread model with the effect of group behavior,and the leader-follower model with the effect of emotion state are proposed.On this basis,exit choice strategies with the effect of emotion state and group behavior are proposed.Fusing emotion spread model,leader-follower model,and exit choice strategies into a cellular automata(CA)-based pedestrian simulation model,we simulate the evacuation process in a multi-exit case.Simulation results indicate that panic emotion and group behavior are two negative influence factors for pedestrian evacuation.Compared with panic emotion or group behavior only,pedestrian evacuation efficiency with the effects of both is lower.

Multilayer Neural Network Based Speech Emotion Recognition for Smart Assistance

Day by day,biometric-based systems play a vital role in our daily lives.This paper proposed an intelligent assistant intended to identify emotions via voice message.A biometric system has been developed to detect human emotions based on voice recognition and control a few electronic peripherals for alert actions.This proposed smart assistant aims to provide a support to the people through buzzer and light emitting diodes(LED)alert signals and it also keep track of the places like households,hospitals and remote areas,etc.The proposed approach is able to detect seven emotions:worry,surprise,neutral,sadness,happiness,hate and love.The key elements for the implementation of speech emotion recognition are voice processing,and once the emotion is recognized,the machine interface automatically detects the actions by buzzer and LED.The proposed system is trained and tested on various benchmark datasets,i.e.,Ryerson Audio-Visual Database of Emotional Speech and Song(RAVDESS)database,Acoustic-Phonetic Continuous Speech Corpus(TIMIT)database,Emotional Speech database(Emo-DB)database and evaluated based on various parameters,i.e.,accuracy,error rate,and time.While comparing with existing technologies,the proposed algorithm gave a better error rate and less time.Error rate and time is decreased by 19.79%,5.13 s.for the RAVDEES dataset,15.77%,0.01 s for the Emo-DB dataset and 14.88%,3.62 for the TIMIT database.The proposed model shows better accuracy of 81.02%for the RAVDEES dataset,84.23%for the TIMIT dataset and 85.12%for the Emo-DB dataset compared to Gaussian Mixture Modeling(GMM)and Support Vector Machine(SVM)Model.

The Efficacy of Deep Learning-Based Mixed Model for Speech Emotion Recognition

Human speech indirectly represents the mental state or emotion of others.The use of Artificial Intelligence(AI)-based techniques may bring revolution in this modern era by recognizing emotion from speech.In this study,we introduced a robust method for emotion recognition from human speech using a well-performed preprocessing technique together with the deep learning-based mixed model consisting of Long Short-Term Memory(LSTM)and Convolutional Neural Network(CNN).About 2800 audio files were extracted from the Toronto emotional speech set(TESS)database for this study.A high pass and Savitzky Golay Filter have been used to obtain noise-free as well as smooth audio data.A total of seven types of emotions;Angry,Disgust,Fear,Happy,Neutral,Pleasant-surprise,and Sad were used in this study.Energy,Fundamental frequency,and Mel Frequency Cepstral Coefficient(MFCC)have been used to extract the emotion features,and these features resulted in 97.5%accuracy in the mixed LSTM+CNN model.This mixed model is found to be performed better than the usual state-of-the-art models in emotion recognition from speech.It also indicates that this mixed model could be effectively utilized in advanced research dealing with sound processing.

A Multi-Level Circulant Cross-Modal Transformer for Multimodal Speech Emotion Recognition

Speech emotion recognition,as an important component of humancomputer interaction technology,has received increasing attention.Recent studies have treated emotion recognition of speech signals as a multimodal task,due to its inclusion of the semantic features of two different modalities,i.e.,audio and text.However,existing methods often fail in effectively represent features and capture correlations.This paper presents a multi-level circulant cross-modal Transformer(MLCCT)formultimodal speech emotion recognition.The proposed model can be divided into three steps,feature extraction,interaction and fusion.Self-supervised embedding models are introduced for feature extraction,which give a more powerful representation of the original data than those using spectrograms or audio features such as Mel-frequency cepstral coefficients(MFCCs)and low-level descriptors(LLDs).In particular,MLCCT contains two types of feature interaction processes,where a bidirectional Long Short-term Memory(Bi-LSTM)with circulant interaction mechanism is proposed for low-level features,while a two-stream residual cross-modal Transformer block is appliedwhen high-level features are involved.Finally,we choose self-attention blocks for fusion and a fully connected layer to make predictions.To evaluate the performance of our proposed model,comprehensive experiments are conducted on three widely used benchmark datasets including IEMOCAP,MELD and CMU-MOSEI.The competitive results verify the effectiveness of our approach.