Despite the remarkable performance and generalization levels of deep learning models in a wide range of artificial intelligence tasks, it has been demonstrated that these models can be easily fooled by the addition of imperceptible yet malicious perturbations to natural inputs. These altered inputs are known in the literature as adversarial examples. In this paper, we propose a novel probabilistic framework to generalize and extend adversarial attacks in order to produce a desired probability distribution for the classes when we apply the attack method to a large number of inputs. This novel attack paradigm provides the adversary with greater control over the target model, thereby exposing, in a wide range of scenarios, threats against deep learning models that cannot be conducted by the conventional paradigms. We introduce four different strategies to efficiently generate such attacks, and illustrate our approach by extending multiple adversarial attack algorithms. We also experimentally validate our approach for the spoken command classification task and the Tweet emotion classification task, two exemplary machine learning problems in the audio and text domain, respectively. Our results demonstrate that we can closely approximate any probability distribution for the classes while maintaining a high fooling rate and even prevent the attacks from being detected by label-shift detection methods.
Emotion intensity prediction determines the degree or intensity of an emotion that the author expresses in a text, extending previous categorical approaches to emotion detection. While most previous work on this topic has concentrated on English texts, other languages would also benefit from fine-grained emotion classification, preferably without having to recreate the amount of annotated data available in English in each new language. Consequently, we explore cross-lingual transfer approaches for fine-grained emotion detection in Spanish and Catalan tweets. To this end we annotate a test set of Spanish and Catalan tweets using Best-Worst scaling. We compare six cross-lingual approaches, e.g., machine translation and cross-lingual embeddings, which have varying requirements for parallel data -- from millions of parallel sentences to completely unsupervised. The results show that on this data, methods with low parallel-data requirements perform surprisingly better than methods that use more parallel data, which we explain through an in-depth error analysis. We make the dataset and the code available at \url{https://github.com/jerbarnes/fine-grained_cross-lingual_emotion}
Automatic emotion categorization has been predominantly formulated as text classification in which textual units are assigned to an emotion from a predefined inventory, for instance following the fundamental emotion classes proposed by Paul Ekman (fear, joy, anger, disgust, sadness, surprise) or Robert Plutchik (adding trust, anticipation). This approach ignores existing psychological theories to some degree, which provide explanations regarding the perception of events. For instance, the description that somebody discovers a snake is associated with fear, based on the appraisal as being an unpleasant and non-controllable situation. This emotion reconstruction is even possible without having access to explicit reports of a subjective feeling (for instance expressing this with the words "I am afraid."). Automatic classification approaches therefore need to learn properties of events as latent variables (for instance that the uncertainty and the mental or physical effort associated with the encounter of a snake leads to fear). With this paper, we propose to make such interpretations of events explicit, following theories of cognitive appraisal of events, and show their potential for emotion classification when being encoded in classification models. Our results show that high quality appraisal dimension assignments in event descriptions lead to an improvement in the classification of discrete emotion categories. We make our corpus of appraisal-annotated emotion-associated event descriptions publicly available.
Most approaches to emotion analysis of social media, literature, news, and other domains focus exclusively on basic emotion categories as defined by Ekman or Plutchik. However, art (such as literature) enables engagement in a broader range of more complex and subtle emotions. These have been shown to also include mixed emotional responses. We consider emotions in poetry as they are elicited in the reader, rather than what is expressed in the text or intended by the author. Thus, we conceptualize a set of aesthetic emotions that are predictive of aesthetic appreciation in the reader, and allow the annotation of multiple labels per line to capture mixed emotions within their context. We evaluate this novel setting in an annotation experiment both with carefully trained experts and via crowdsourcing. Our annotation with experts leads to an acceptable agreement of kappa = .70, resulting in a consistent dataset for future large scale analysis. Finally, we conduct first emotion classification experiments based on BERT, showing that identifying aesthetic emotions is challenging in our data, with up to .52 F1-micro on the German subset. Data and resources are available at https://github.com/tnhaider/poetry-emotion
Emotion cause analysis such as emotion cause extraction (ECE) and emotion-cause pair extraction (ECPE) have gradually attracted the attention of many researchers. However, there are still two shortcomings in the existing research: 1) In most cases, emotion expression and cause are not the whole clause, but the span in the clause, so extracting the clause-pair rather than the span-pair greatly limits its applications in real-world scenarios; 2) It is not enough to extract the emotion expression clause without identifying the emotion categories, the presence of emotion clause does not necessarily convey emotional information explicitly due to different possible causes. In this paper, we propose a new task: Emotion-Cause Span-Pair extraction and classification (ECSP), which aims to extract the potential span-pair of emotion and corresponding causes in a document, and make emotion classification for each pair. In the new ECSP task, ECE and ECPE can be regarded as two special cases at the clause-level. We propose a span-based extract-then-classify (ETC) model, where emotion and cause are directly extracted and paired from the document under the supervision of target span boundaries, and corresponding categories are then classified using their pair representations and localized context. Experiments show that our proposed ETC model outperforms the SOTA model of ECE and ECPE task respectively and gets a fair-enough results on ECSP task.
We present ProxEmo, a novel end-to-end emotion prediction algorithm for socially aware robot navigation among pedestrians. Our approach predicts the perceived emotions of a pedestrian from walking gaits, which is then used for emotion-guided navigation taking into account social and proxemic constraints. To classify emotions, we propose a multi-view skeleton graph convolution-based model that works on a commodity camera mounted onto a moving robot. Our emotion recognition is integrated into a mapless navigation scheme and makes no assumptions about the environment of pedestrian motion. It achieves a mean average emotion prediction precision of 82.47% on the Emotion-Gait benchmark dataset. We outperform current state-of-art algorithms for emotion recognition from 3D gaits. We highlight its benefits in terms of navigation in indoor scenes using a Clearpath Jackal robot.
In this work, we explore the dependencies between speaker recognition and emotion recognition. We first show that knowledge learned for speaker recognition can be reused for emotion recognition through transfer learning. Then, we show the effect of emotion on speaker recognition. For emotion recognition, we show that using a simple linear model is enough to obtain good performance on the features extracted from pre-trained models such as the x-vector model. Then, we improve emotion recognition performance by fine-tuning for emotion classification. We evaluated our experiments on three different types of datasets: IEMOCAP, MSP-Podcast, and Crema-D. By fine-tuning, we obtained 30.40%, 7.99%, and 8.61% absolute improvement on IEMOCAP, MSP-Podcast, and Crema-D respectively over baseline model with no pre-training. Finally, we present results on the effect of emotion on speaker verification. We observed that speaker verification performance is prone to changes in test speaker emotions. We found that trials with angry utterances performed worst in all three datasets. We hope our analysis will initiate a new line of research in the speaker recognition community.
This paper presents a neural network based method Multi-Task Affect Net(MTANet) submitted to the Affective Behavior Analysis in-the-Wild Challenge in FG2020. This method is a multi-task network and based on SE-ResNet modules. By utilizing multi-task learning, this network can estimate and recognize three quantified affective models: valence and arousal, action units, and seven basic emotions simultaneously. MTANet achieve Concordance Correlation Coefficient(CCC) rates of 0.28 and 0.34 for valence and arousal, F1-score of 0.427 and 0.32 for AUs detection and categorical emotion classification.
Image and video-capturing technologies have permeated our every-day life. Such technologies can continuously monitor individuals' expressions in real-life settings, affording us new insights into their emotional states and transitions, thus paving the way to novel well-being and healthcare applications. Yet, due to the strong privacy concerns, the use of such technologies is met with strong skepticism, since current face-based emotion recognition systems relying on deep learning techniques tend to preserve substantial information related to the identity of the user, apart from the emotion-specific information. This paper proposes an adversarial learning framework which relies on a convolutional neural network (CNN) architecture trained through an iterative procedure for minimizing identity-specific information and maximizing emotion-dependent information. The proposed approach is evaluated through emotion classification and face identification metrics, and is compared against two CNNs, one trained solely for emotion recognition and the other trained solely for face identification. Experiments are performed using the Yale Face Dataset and Japanese Female Facial Expression Database. Results indicate that the proposed approach can learn a convolutional transformation for preserving emotion recognition accuracy and degrading face identity recognition, providing a foundation toward privacy-aware emotion recognition technologies.
To enable humanoid robots to share our social space we need to develop technology for easy interaction with the robots using multiple modes such as speech, gestures and share our emotions with them. We have targeted this research towards addressing the core issue of emotion recognition problem which would require less computation resources and much lesser number of network hyperparameters which will be more adaptive to be computed on low resourced social robots for real time communication. More specifically, here we have proposed an Inception module based Convolutional Neural Network Architecture which has achieved improved accuracy of upto 6% improvement over the existing network architecture for emotion classification when combinedly tested over multiple datasets when tried over humanoid robots in real - time. Our proposed model is reducing the trainable Hyperparameters to an extent of 94% as compared to vanilla CNN model which clearly indicates that it can be used in real time based application such as human robot interaction. Rigorous experiments have been performed to validate our methodology which is sufficiently robust and could achieve high level of accuracy. Finally, the model is implemented in a humanoid robot, NAO in real time and robustness of the model is evaluated.
Physiological signals that provide the objective repression of human affective states are attracted increasing attention in the emotion recognition field. However, the single signal is difficult to obtain completely and accurately description for emotion. Multiple physiological signals fusing models, building the uniform classification model by means of consistent and complementary information from different emotions to improve recognition performance. Original fusing models usually choose the particular classification method to recognition, which is ignoring different distribution of multiple signals. Aiming above problems, in this work, we propose an emotion classification model through multiple modal physiological signals for different emotions. Features are extracted from EEG, EMG, EOG signals for characterizing emotional state on valence and arousal levels. For characterization, four bands filtering theta, beta, alpha, gamma for signal preprocessing are adopted and three Hjorth parameters are computing as features. To improve classification performance, an ensemble classifier is built. Experiments are conducted on the benchmark DEAP datasets. For the two-class task, the best result on arousal is 94.42\%, the best result on valence is 94.02\%, respectively. For the four-class task, the highest average classification accuracy is 90.74, and it shows good stability. The influence of different peripheral physiological signals for results is also analyzed in this paper.
In this work, we conduct an extensive comparison of various approaches to speech based emotion recognition systems. The analyses were carried out on audio recordings from Ryerson Audio-Visual Database of Emotional Speech and Song (RAVDESS). After pre-processing the raw audio files, features such as Log-Mel Spectrogram, Mel-Frequency Cepstral Coefficients (MFCCs), pitch and energy were considered. The significance of these features for emotion classification was compared by applying methods such as Long Short Term Memory (LSTM), Convolutional Neural Networks (CNNs), Hidden Markov Models (HMMs) and Deep Neural Networks (DNNs). On the 14-class (2 genders x 7 emotions) classification task, an accuracy of 68% was achieved with a 4-layer 2 dimensional CNN using the Log-Mel Spectrogram features. We also observe that, in emotion recognition, the choice of audio features impacts the results much more than the model complexity.
Most research on emotion analysis from text focuses on the task of emotion classification or emotion intensity regression. Fewer works address emotions as a phenomenon to be tackled with structured learning, which can be explained by the lack of relevant datasets. We fill this gap by releasing a dataset of 5000 English news headlines annotated via crowdsourcing with their associated emotions, the corresponding emotion experiencers and textual cues, related emotion causes and targets, as well as the reader's perception of the emotion of the headline. This annotation task is comparably challenging, given the large number of classes and roles to be identified. We therefore propose a multiphase annotation procedure in which we first find relevant instances with emotional content and then annotate the more fine-grained aspects. Finally, we develop a baseline for the task of automatic prediction of semantic role structures and discuss the results. The corpus we release enables further research on emotion classification, emotion intensity prediction, emotion cause detection, and supports further qualitative studies.
Microblogs are widely used to express people's opinions and feelings in daily life. Sentiment analysis (SA) can timely detect personal sentiment polarities through analyzing text. Deep learning approaches have been broadly used in SA but still have not fully exploited syntax information. In this paper, we propose a syntax-based graph convolution network (GCN) model to enhance the understanding of diverse grammatical structures of Chinese microblogs. In addition, a pooling method based on percentile is proposed to improve the accuracy of the model. In experiments, for Chinese microblogs emotion classification categories including happiness, sadness, like, anger, disgust, fear, and surprise, the F-measure of our model reaches 82.32% and exceeds the state-of-the-art algorithm by 5.90%. The experimental results show that our model can effectively utilize the information of dependency parsing to improve the performance of emotion detection. What is more, we annotate a new dataset for Chinese emotion classification, which is open to other researchers.
In this work, we explore the impact of visual modality in addition to speech and text for improving the accuracy of the emotion detection system. The traditional approaches tackle this task by fusing the knowledge from the various modalities independently for performing emotion classification. In contrast to these approaches, we tackle the problem by introducing an attention mechanism to combine the information. In this regard, we first apply a neural network to obtain hidden representations of the modalities. Then, the attention mechanism is defined to select and aggregate important parts of the video data by conditioning on the audio and text data. Furthermore, the attention mechanism is again applied to attend important parts of the speech and textual data, by considering other modality. Experiments are performed on the standard IEMOCAP dataset using all three modalities (audio, text, and video). The achieved results show a significant improvement of 3.65% in terms of weighted accuracy compared to the baseline system.
The recognition of emotion and dialogue acts enriches conversational analysis and help to build natural dialogue systems. Emotion interpretation makes us understand feelings and dialogue acts reflect the intentions and performative functions in the utterances. However, most of the textual and multi-modal conversational emotion corpora contain only emotion labels but not dialogue acts. To address this problem, we propose to use a pool of various recurrent neural models trained on a dialogue act corpus, with and without context. These neural models annotate the emotion corpora with dialogue act labels, and an ensemble annotator extracts the final dialogue act label. We annotated two accessible multi-modal emotion corpora: IEMOCAP and MELD. We analyzed the co-occurrence of emotion and dialogue act labels and discovered specific relations. For example, Accept/Agree dialogue acts often occur with the Joy emotion, Apology with Sadness, and Thanking with Joy. We make the Emotional Dialogue Acts (EDA) corpus publicly available to the research community for further study and analysis.
Can vocal emotions be emulated? This question has been a recurrent concern of the speech community, and has also been vigorously investigated. It has been fueled further by its link to the issue of validity of acted emotion databases. Much of the speech and vocal emotion research has relied on acted emotion databases as valid proxies for studying natural emotions. To create models that generalize to natural settings, it is crucial to work with valid prototypes -- ones that can be assumed to reliably represent natural emotions. More concretely, it is important to study emulated emotions against natural emotions in terms of their physiological, and psychological concomitants. In this paper, we present an on-scale systematic study of the differences between natural and acted vocal emotions. We use a self-attention based emotion classification model to understand the phonetic bases of emotions by discovering the most 'attended' phonemes for each class of emotions. We then compare these attended-phonemes in their importance and distribution across acted and natural classes. Our tests show significant differences in the manner and choice of phonemes in acted and natural speech, concluding moderate to low validity and value in using acted speech databases for emotion classification tasks.
We present a model to predict fine-grained emotions along the continuous dimensions of valence, arousal, and dominance (VAD) with a corpus with categorical emotion annotations. Our model is trained by minimizing the EMD (Earth Mover's Distance) loss between the predicted VAD score distribution and the categorical emotion distributions sorted along VAD, and it can simultaneously classify the emotion categories and predict the VAD scores for a given sentence. We use pre-trained RoBERTa-Large and fine-tune on three different corpora with categorical labels and evaluate on EmoBank corpus with VAD scores. We show that our approach reaches comparable performance to that of the state-of-the-art classifiers in categorical emotion classification and shows significant positive correlations with the ground truth VAD scores. Also, further training with supervision of VAD labels leads to improved performance especially when dataset is small. We also present examples of predictions of appropriate emotion words that are not part of the original annotations.
Emotion detection in text is an important task in NLP and is essential in many applications. Most of the existing methods treat this task as a problem of single-label multi-class text classification. To predict multiple emotions for one instance, most of the existing works regard it as a general Multi-label Classification (MLC) problem, where they usually either apply a manually determined threshold on the last output layer of their neural network models or train multiple binary classifiers and make predictions in the fashion of one-vs-all. However, compared to labels in the general MLC datasets, the number of emotion categories are much fewer (less than 10). Additionally, emotions tend to have more correlations with each other. For example, the human usually does not express "joy" and "anger" at the same time, but it is very likely to have "joy" and "love" expressed together. Given this intuition, in this paper, we propose a Latent Variable Chain (LVC) transformation and a tailored model -- Seq2Emo model that not only naturally predicts multiple emotion labels but also takes into consideration their correlations. We perform the experiments on the existing multi-label emotion datasets as well as on our newly collected datasets. The results show that our model compares favorably with existing state-of-the-art methods.
Representation learning for speech emotion recognition is challenging due to labeled data sparsity issue and lack of gold standard references. In addition, there is much variability from input speech signals, human subjective perception of the signals and emotion label ambiguity. In this paper, we propose a machine learning framework to obtain speech emotion representations by limiting the effect of speaker variability in the speech signals. Specifically, we propose to disentangle the speaker characteristics from emotion through an adversarial training network in order to better represent emotion. Our method combines the gradient reversal technique with an entropy loss function to remove such speaker information. Our approach is evaluated on both IEMOCAP and CMU-MOSEI datasets. We show that our method improves speech emotion classification and increases generalization to unseen speakers.