Electroencephalography (EEG) is shown to be a valuable data source for evaluating subjects' mental states. However, the interpretation of multi-modal EEG signals is challenging, as they suffer from poor signal-to-noise-ratio, are highly subject-dependent, and are bound to the equipment and experimental setup used, (i.e. domain). This leads to machine learning models often suffer from poor generalization ability, where they perform significantly worse on real-world data than on the exploited training data. Recent research heavily focuses on cross-subject and cross-session transfer learning frameworks to reduce domain calibration efforts for EEG signals. We argue that multi-source learning via learning domain-invariant representations from multiple data-sources is a viable alternative, as the available data from different EEG data-source domains (e.g., subjects, sessions, experimental setups) grow massively. We propose an adversarial inference approach to learn data-source invariant representations in this context, enabling multi-source learning for EEG-based brain-computer interfaces. We unify EEG recordings from different source domains (i.e., emotion recognition datasets SEED, SEED-IV, DEAP, DREAMER), and demonstrate the feasibility of our invariant representation learning approach in suppressing data-source-relevant information leakage by 35% while still achieving stable EEG-based emotion classification performance.
Speech emotion recognition~(SER) refers to the technique of inferring the emotional state of an individual from speech signals. SERs continue to garner interest due to their wide applicability. Although the domain is mainly founded on signal processing, machine learning, and deep learning, generalizing over languages continues to remain a challenge. However, developing generalizable and transferable models are critical due to a lack of sufficient resources in terms of data and labels for languages beyond the most commonly spoken ones. To improve performance over languages, we propose a denoising autoencoder with semi-supervision using a continuous metric loss based on either activation or valence. The novelty of this work lies in our proposal of continuous metric learning, which is among the first proposals on the topic to the best of our knowledge. Furthermore, to address the lack of activation and valence labels in the transfer datasets, we annotate the signal samples with activation and valence levels corresponding to a dimensional model of emotions, which were then used to evaluate the quality of the embedding over the transfer datasets. We show that the proposed semi-supervised model consistently outperforms the baseline unsupervised method, which is a conventional denoising autoencoder, in terms of emotion classification accuracy as well as correlation with respect to the dimensional variables. Further evaluation of classification accuracy with respect to the reference, a BERT based speech representation model, shows that the proposed method is comparable to the reference method in classifying specific emotion classes at a much lower complexity.
In recent years, speech emotion recognition (SER) has been used in wide ranging applications, from healthcare to the commercial sector. In addition to signal processing approaches, methods for SER now also use deep learning techniques which provide transfer learning possibilities. However, generalizing over languages, corpora and recording conditions is still an open challenge. In this work we address this gap by exploring loss functions that aid in transferability, specifically to non-tonal languages. We propose a variational autoencoder (VAE) with KL annealing and a semi-supervised VAE to obtain more consistent latent embedding distributions across data sets. To ensure transferability, the distribution of the latent embedding should be similar across non-tonal languages (data sets). We start by presenting a low-complexity SER based on a denoising-autoencoder, which achieves an unweighted classification accuracy of over 52.09% for four-class emotion classification. This performance is comparable to that of similar baseline methods. Following this, we employ a VAE, the semi-supervised VAE and the VAE with KL annealing to obtain a more regularized latent space. We show that while the DAE has the highest classification accuracy among the methods, the semi-supervised VAE has a comparable classification accuracy and a more consistent latent embedding distribution over data sets.
Emotion recognition in conversation (ERC) aims to analyze the speaker's state and identify their emotion in the conversation. Recent works in ERC focus on context modeling but ignore the representation of contextual emotional tendency. In order to extract multi-modal information and the emotional tendency of the utterance effectively, we propose a new structure named Emoformer to extract multi-modal emotion vectors from different modalities and fuse them with sentence vector to be an emotion capsule. Furthermore, we design an end-to-end ERC model called EmoCaps, which extracts emotion vectors through the Emoformer structure and obtain the emotion classification results from a context analysis model. Through the experiments with two benchmark datasets, our model shows better performance than the existing state-of-the-art models.
Emotion classification is often formulated as the task to categorize texts into a predefined set of emotion classes. So far, this task has been the recognition of the emotion of writers and readers, as well as that of entities mentioned in the text. We argue that a classification setup for emotion analysis should be performed in an integrated manner, including the different semantic roles that participate in an emotion episode. Based on appraisal theories in psychology, which treat emotions as reactions to events, we compile an English corpus of written event descriptions. The descriptions depict emotion-eliciting circumstances, and they contain mentions of people who responded emotionally. We annotate all experiencers, including the original author, with the emotions they likely felt. In addition, we link them to the event they found salient (which can be different for different experiencers in a text) by annotating event properties, or appraisals (e.g., the perceived event undesirability, the uncertainty of its outcome). Our analysis reveals patterns in the co-occurrence of people's emotions in interaction. Hence, this richly-annotated resource provides useful data to study emotions and event evaluations from the perspective of different roles, and it enables the development of experiencer-specific emotion and appraisal classification systems.
Emotional recognition through exploring the electroencephalography (EEG) characteristics has been widely performed in recent studies. Nonlinear analysis and feature extraction methods for understanding the complex dynamical phenomena are associated with the EEG patterns of different emotions. The phase space reconstruction is a typical nonlinear technique to reveal the dynamics of the brain neural system. Recently, the topological data analysis (TDA) scheme has been used to explore the properties of space, which provides a powerful tool to think over the phase space. In this work, we proposed a topological EEG nonlinear dynamics analysis approach using the phase space reconstruction (PSR) technique to convert EEG time series into phase space, and the persistent homology tool explores the topological properties of the phase space. We perform the topological analysis of EEG signals in different rhythm bands to build emotion feature vectors, which shows high distinguishing ability. We evaluate the approach with two well-known benchmark datasets, the DEAP and DREAMER datasets. The recognition results achieved accuracies of 99.37% and 99.35% in arousal and valence classification tasks with DEAP, and 99.96%, 99.93%, and 99.95% in arousal, valence, and dominance classifications tasks with DREAMER, respectively. The performances are supposed to be outperformed current state-of-art approaches in DREAMER (improved by 1% to 10% depends on temporal length), while comparable to other related works evaluated in DEAP. The proposed work is the first investigation in the emotion recognition oriented EEG topological feature analysis, which brought a novel insight into the brain neural system nonlinear dynamics analysis and feature extraction.
Emotion recognition is a key attribute for artificial intelligence systems that need to naturally interact with humans. However, the task definition is still an open problem due to the inherent ambiguity of emotions. In this paper, a novel Bayesian training loss based on per-utterance Dirichlet prior distributions is proposed for verbal emotion recognition, which models the uncertainty in one-hot labels created when human annotators assign the same utterance to different emotion classes. An additional metric is used to evaluate the performance by detection test utterances with high labelling uncertainty. This removes a major limitation that emotion classification systems only consider utterances with labels where the majority of annotators agree on the emotion class. Furthermore, a frequentist approach is studied to leverage the continuous-valued "soft" labels obtained by averaging the one-hot labels. We propose a two-branch model structure for emotion classification on a per-utterance basis, which achieves state-of-the-art classification results on the widely used IEMOCAP dataset. Based on this, uncertainty estimation experiments were performed. The best performance in terms of the area under the precision-recall curve when detecting utterances with high uncertainty was achieved by interpolating the Bayesian training loss with the Kullback-Leibler divergence training loss for the soft labels. The generality of the proposed approach was verified using the MSP-Podcast dataset which yielded the same pattern of results.
Accurate emotion classification for online reviews is vital for business organizations to gain deeper insights into markets. Although deep learning has been successfully implemented in this area, accuracy and processing time are still major problems preventing it from reaching its full potential. This paper proposes an Enhanced Leaky Rectified Linear Unit activation and Weighted Loss (ELReLUWL) algorithm for enhanced text emotion classification and faster parameter convergence speed. This algorithm includes the definition of the inflection point and the slope for inputs on the left side of the inflection point to avoid gradient saturation. It also considers the weight of samples belonging to each class to compensate for the influence of data imbalance. Convolutional Neural Network (CNN) combined with the proposed algorithm to increase the classification accuracy and decrease the processing time by eliminating the gradient saturation problem and minimizing the negative effect of data imbalance, demonstrated on a binary sentiment problem. The results show that the proposed solution achieves better classification performance in different data scenarios and different review types. The proposed model takes less convergence time to achieve model optimization with seven epochs against the current convergence time of 11.5 epochs on average. The proposed solution improves accuracy and reduces the processing time of text emotion classification. The solution provides an average class accuracy of 96.63% against a current average accuracy of 91.56%. It also provides a processing time of 23.3 milliseconds compared to the current average processing time of 33.2 milliseconds. Finally, this study solves the issues of gradient saturation and data imbalance. It enhances overall average class accuracy and decreases processing time.
In recent years, Internet memes have been widely used in online chatting. Compared with text-based communication, conversations become more expressive and attractive when Internet memes are incorporated. This paper presents our solutions for the Meme incorporated Open-domain Dialogue (MOD) Challenge of DSTC10, where three tasks are involved: text response modeling, meme retrieval, and meme emotion classification. Firstly, we leverage a large-scale pre-trained dialogue model for coherent and informative response generation. Secondly, based on interaction-based text-matching, our approach can retrieve appropriate memes with good generalization ability. Thirdly, we propose to model the emotion flow (EF) in conversations and introduce an auxiliary task of emotion description prediction (EDP) to boost the performance of meme emotion classification. Experimental results on the MOD dataset demonstrate that our methods can incorporate Internet memes into dialogue systems effectively.
Humans express their emotions via facial expressions, voice intonation and word choices. To infer the nature of the underlying emotion, recognition models may use a single modality, such as vision, audio, and text, or a combination of modalities. Generally, models that fuse complementary information from multiple modalities outperform their uni-modal counterparts. However, a successful model that fuses modalities requires components that can effectively aggregate task-relevant information from each modality. As cross-modal attention is seen as an effective mechanism for multi-modal fusion, in this paper we quantify the gain that such a mechanism brings compared to the corresponding self-attention mechanism. To this end, we implement and compare a cross-attention and a self-attention model. In addition to attention, each model uses convolutional layers for local feature extraction and recurrent layers for global sequential modelling. We compare the models using different modality combinations for a 7-class emotion classification task using the IEMOCAP dataset. Experimental results indicate that albeit both models improve upon the state-of-the-art in terms of weighted and unweighted accuracy for tri- and bi-modal configurations, their performance is generally statistically comparable. The code to replicate the experiments is available at https://github.com/smartcameras/SelfCrossAttn
We propose PARSE, a novel semi-supervised architecture for learning strong EEG representations for emotion recognition. To reduce the potential distribution mismatch between the large amounts of unlabeled data and the limited amount of labeled data, PARSE uses pairwise representation alignment. First, our model performs data augmentation followed by label guessing for large amounts of original and augmented unlabeled data. This is then followed by sharpening of the guessed labels and convex combinations of the unlabeled and labeled data. Finally, representation alignment and emotion classification are performed. To rigorously test our model, we compare PARSE to several state-of-the-art semi-supervised approaches which we implement and adapt for EEG learning. We perform these experiments on four public EEG-based emotion recognition datasets, SEED, SEED-IV, SEED-V and AMIGOS (valence and arousal). The experiments show that our proposed framework achieves the overall best results with varying amounts of limited labeled samples in SEED, SEED-IV and AMIGOS (valence), while approaching the overall best result (reaching the second-best) in SEED-V and AMIGOS (arousal). The analysis shows that our pairwise representation alignment considerably improves the performance by reducing the distribution alignment between unlabeled and labeled data, especially when only 1 sample per class is labeled.
Electroencephalogram (EEG) is one of the most reliable physiological signal for emotion detection. Being non-stationary in nature, EEGs are better analysed by spectro temporal representations. Standard features like Discrete Wavelet Transformation (DWT) can represent temporal changes in spectral dynamics of an EEG, but is insufficient to extract information other way around, i.e. spectral changes in temporal dynamics. On the other hand, Empirical mode decomposition (EMD) based features can be useful to bridge the above mentioned gap. Towards this direction, we extract two novel features on top of EMD, namely, (a) marginal hilbert spectrum (MHS) and (b) Holo-Hilbert spectral analysis (HHSA) based on EMD, to better represent emotions in 2D arousal-valence (A-V) space. The usefulness of these features for EEG emotion classification is investigated through extensive experiments using state-of-the-art classifiers. In addition, experiments conducted on DEAP dataset for binary emotion classification in both A-V space, reveal the efficacy of the proposed features over the standard set of temporal and spectral features.
Facial emotion recognition is a vast and complex problem space within the domain of computer vision and thus requires a universally accepted baseline method with which to evaluate proposed models. While test datasets have served this purpose in the academic sphere real world application and testing of such models lacks any real comparison. Therefore we propose a framework in which models developed for FER can be compared and contrasted against one another in a constant standardized fashion. A lightweight convolutional neural network is trained on the AffectNet dataset a large variable dataset for facial emotion recognition and a web application is developed and deployed with our proposed framework as a proof of concept. The CNN is embedded into our application and is capable of instant real time facial emotion recognition. When tested on the AffectNet test set this model achieves high accuracy for emotion classification of eight different emotions. Using our framework the validity of this model and others can be properly tested by evaluating a model efficacy not only based on its accuracy on a sample test dataset, but also on in the wild experiments. Additionally, our application is built with the ability to save and store any image captured or uploaded to it for emotion recognition, allowing for the curation of more quality and diverse facial emotion recognition datasets.
Emotion recognition using EEG has been widely studied to address the challenges associated with affective computing. Using manual feature extraction methods on EEG signals results in sub-optimal performance by the learning models. With the advancements in deep learning as a tool for automated feature engineering, in this work, a hybrid of manual and automatic feature extraction methods has been proposed. The asymmetry in different brain regions is captured in a 2D vector, termed the AsMap, from the differential entropy features of EEG signals. These AsMaps are then used to extract features automatically using a convolutional neural network model. The proposed feature extraction method has been compared with differential entropy and other feature extraction methods such as relative asymmetry, differential asymmetry and differential caudality. Experiments are conducted using the SJTU emotion EEG dataset and the DEAP dataset on different classification problems based on the number of classes. Results obtained indicate that the proposed method of feature extraction results in higher classification accuracy, outperforming the other feature extraction methods. The highest classification accuracy of 97.10% is achieved on a three-class classification problem using the SJTU emotion EEG dataset. Further, this work has also assessed the impact of window size on classification accuracy.
Deep learning based electroencephalography (EEG) signal processing methods are known to suffer from poor test-time generalization due to the changes in data distribution. This becomes a more challenging problem when privacy-preserving representation learning is of interest such as in clinical settings. To that end, we propose a multi-source learning architecture where we extract domain-invariant representations from dataset-specific private encoders. Our model utilizes a maximum-mean-discrepancy (MMD) based domain alignment approach to impose domain-invariance for encoded representations, which outperforms state-of-the-art approaches in EEG-based emotion classification. Furthermore, representations learned in our pipeline preserve domain privacy as dataset-specific private encoding alleviates the need for conventional, centralized EEG-based deep neural network training approaches with shared parameters.
Contrastive Language-Image Pre-training (CLIP) represents the latest incarnation of pre-trained vision-language models. Although CLIP has recently shown its superior power on a wide range of downstream vision-language tasks like Visual Question Answering, it is still underexplored for Image Emotion Classification (IEC). Adapting CLIP to the IEC task has three significant challenges, tremendous training objective gap between pretraining and IEC, shared suboptimal and invariant prompts for all instances. In this paper, we propose a general framework that shows how CLIP can be effectively applied to IEC. We first introduce a prompt tuning method that mimics the pretraining objective of CLIP and thus can leverage the rich image and text semantics entailed in CLIP. Then we automatically compose instance-specific prompts by conditioning them on the categories and image contents of instances, diversifying prompts and avoiding suboptimal problems. Evaluations on six widely-used affective datasets demonstrate that our proposed method outperforms the state-of-the-art methods to a large margin (i.e., up to 9.29% accuracy gain on EmotionROI dataset) on IEC tasks, with only a few parameters trained. Our codes will be publicly available for research purposes.
In this paper, we introduce a pretrained audio-visual Transformer trained on more than 500k utterances from nearly 4000 celebrities from the VoxCeleb2 dataset for human behavior understanding. The model aims to capture and extract useful information from the interactions between human facial and auditory behaviors, with application in emotion recognition. We evaluate the model performance on two datasets, namely CREMAD-D (emotion classification) and MSP-IMPROV (continuous emotion regression). Experimental results show that fine-tuning the pre-trained model helps improving emotion classification accuracy by 5-7% and Concordance Correlation Coefficients (CCC) in continuous emotion recognition by 0.03-0.09 compared to the same model trained from scratch. We also demonstrate the robustness of finetuning the pre-trained model in a low-resource setting. With only 10% of the original training set provided, fine-tuning the pre-trained model can lead to at least 10% better emotion recognition accuracy and a CCC score improvement by at least 0.1 for continuous emotion recognition.
Emotional voice conversion (EVC) seeks to convert the emotional state of an utterance while preserving the linguistic content and speaker identity. In EVC, emotions are usually treated as discrete categories overlooking the fact that speech also conveys emotions with various intensity levels that the listener can perceive. In this paper, we aim to explicitly characterize and control the intensity of emotion. We propose to disentangle the speaker style from linguistic content and encode the speaker style into a style embedding in a continuous space that forms the prototype of emotion embedding. We further learn the actual emotion encoder from an emotion-labelled database and study the use of relative attributes to represent fine-grained emotion intensity. To ensure emotional intelligibility, we incorporate emotion classification loss and emotion embedding similarity loss into the training of the EVC network. As desired, the proposed network controls the fine-grained emotion intensity in the output speech. Through both objective and subjective evaluations, we validate the effectiveness of the proposed network for emotional expressiveness and emotion intensity control.
EEG signals in emotion recognition absorb special attention owing to their high temporal resolution and their information about what happens in the brain. Different regions of brain work together to process information and meanwhile the activity of brain changes during the time. Therefore, the investigation of the connection between different brain areas and their temporal patterns plays an important role in neuroscience. In this study, we investigate the emotion classification performance using functional connectivity features in different frequency bands and compare them with the classification performance using differential entropy feature, which has been previously used for this task. Moreover, we investigate the effect of using different time periods on the classification performance. Our results on publicly available SEED dataset show that as time goes on, emotions become more stable and the classification accuracy increases. Among different time periods, we achieve the highest classification accuracy using the time period of 140s-end. In this time period, the accuracy is improved by 4 to 6% compared to using the entire signal. The mean accuracy of about 88% is obtained using any of the Pearson correlation coefficient, coherence, and phase locking value features and SVM. Therefore, functional connectivity features lead to better classification accuracy than DE features (with the mean accuracy of 84.89%) using the proposed framework. Finally, in a relatively fair comparison, we show that using the best time interval and SVM, we achieve better accuracy than using Recurrent Neural Networks which need large amount of data and have high computational cost.
The portrayal of negative emotions such as anger can vary widely between cultures and contexts, depending on the acceptability of expressing full-blown emotions rather than suppression to maintain harmony. The majority of emotional datasets collect data under the broad label ``anger", but social signals can range from annoyed, contemptuous, angry, furious, hateful, and more. In this work, we curated the first in-the-wild multicultural video dataset of emotions, and deeply explored anger-related emotional expressions by asking culture-fluent annotators to label the videos with 6 labels and 13 emojis in a multi-label framework. We provide a baseline multi-label classifier on our dataset, and show how emojis can be effectively used as a language-agnostic tool for annotation.