Patrick Cardinal

R. Gnana Praveen, Wheidima Carneiro de Melo, Nasib Ullah et al.

Multimodal emotion recognition has recently gained much attention since it can leverage diverse and complementary relationships over multiple modalities (e.g., audio, visual, biosignals, etc.), and can provide some robustness to noisy modalities. Most state-of-the-art methods for audio-visual (A-V) fusion rely on recurrent networks or conventional attention mechanisms that do not effectively leverage the complementary nature of A-V modalities. In this paper, we focus on dimensional emotion recognition based on the fusion of facial and vocal modalities extracted from videos. Specifically, we propose a joint cross-attention model that relies on the complementary relationships to extract the salient features across A-V modalities, allowing for accurate prediction of continuous values of valence and arousal. The proposed fusion model efficiently leverages the inter-modal relationships, while reducing the heterogeneity between the features. In particular, it computes the cross-attention weights based on correlation between the combined feature representation and individual modalities. By deploying the combined A-V feature representation into the cross-attention module, the performance of our fusion module improves significantly over the vanilla cross-attention module. Experimental results on validation-set videos from the AffWild2 dataset indicate that our proposed A-V fusion model provides a cost-effective solution that can outperform state-of-the-art approaches. The code is available on GitHub: https://github.com/praveena2j/JointCrossAttentional-AV-Fusion.

R Gnana Praveen, Eric Granger, Patrick Cardinal

Automatic emotion recognition (ER) has recently gained lot of interest due to its potential in many real-world applications. In this context, multimodal approaches have been shown to improve performance (over unimodal approaches) by combining diverse and complementary sources of information, providing some robustness to noisy and missing modalities. In this paper, we focus on dimensional ER based on the fusion of facial and vocal modalities extracted from videos, where complementary audio-visual (A-V) relationships are explored to predict an individual's emotional states in valence-arousal space. Most state-of-the-art fusion techniques rely on recurrent networks or conventional attention mechanisms that do not effectively leverage the complementary nature of A-V modalities. To address this problem, we introduce a joint cross-attentional model for A-V fusion that extracts the salient features across A-V modalities, that allows to effectively leverage the inter-modal relationships, while retaining the intra-modal relationships. In particular, it computes the cross-attention weights based on correlation between the joint feature representation and that of the individual modalities. By deploying the joint A-V feature representation into the cross-attention module, it helps to simultaneously leverage both the intra and inter modal relationships, thereby significantly improving the performance of the system over the vanilla cross-attention module. The effectiveness of our proposed approach is validated experimentally on challenging videos from the RECOLA and AffWild2 datasets. Results indicate that our joint cross-attentional A-V fusion model provides a cost-effective solution that can outperform state-of-the-art approaches, even when the modalities are noisy or absent.

R Gnana Praveen, Eric Granger, Patrick Cardinal

In video-based emotion recognition (ER), it is important to effectively leverage the complementary relationship among audio (A) and visual (V) modalities, while retaining the intra-modal characteristics of individual modalities. In this paper, a recursive joint attention model is proposed along with long short-term memory (LSTM) modules for the fusion of vocal and facial expressions in regression-based ER. Specifically, we investigated the possibility of exploiting the complementary nature of A and V modalities using a joint cross-attention model in a recursive fashion with LSTMs to capture the intra-modal temporal dependencies within the same modalities as well as among the A-V feature representations. By integrating LSTMs with recursive joint cross-attention, our model can efficiently leverage both intra- and inter-modal relationships for the fusion of A and V modalities. The results of extensive experiments performed on the challenging Affwild2 and Fatigue (private) datasets indicate that the proposed A-V fusion model can significantly outperform state-of-art-methods.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

This paper investigates the impact of different standard environmental sound representations (spectrograms) on the recognition performance and adversarial attack robustness of a victim residual convolutional neural network, namely ResNet-18. Our main motivation for focusing on such a front-end classifier rather than other complex architectures is balancing recognition accuracy and the total number of training parameters. Herein, we measure the impact of different settings required for generating more informative Mel-frequency cepstral coefficient (MFCC), short-time Fourier transform (STFT), and discrete wavelet transform (DWT) representations on our front-end model. This measurement involves comparing the classification performance over the adversarial robustness. We demonstrate an inverse relationship between recognition accuracy and model robustness against six benchmarking attack algorithms on the balance of average budgets allocated by the adversary and the attack cost. Moreover, our experimental results have shown that while the ResNet-18 model trained on DWT spectrograms achieves a high recognition accuracy, attacking this model is relatively more costly for the adversary than other 2D representations. We also report some results on different convolutional neural network architectures such as ResNet-34, ResNet-56, AlexNet, and GoogLeNet, SB-CNN, and LSTM-based.

Mohammad Esmaeilpour, Nourhene Chaalia, Patrick Cardinal

This paper introduces a new synthesis-based defense algorithm for counteracting with a varieties of adversarial attacks developed for challenging the performance of the cutting-edge speech-to-text transcription systems. Our algorithm implements a Sobolev-based GAN and proposes a novel regularizer for effectively controlling over the functionality of the entire generative model, particularly the discriminator network during training. Our achieved results upon carrying out numerous experiments on the victim DeepSpeech, Kaldi, and Lingvo speech transcription systems corroborate the remarkable performance of our defense approach against a comprehensive range of targeted and non-targeted adversarial attacks.

Mohammad Esmaeilpour, Nourhene Chaalia, Adel Abusitta et al.

This paper introduces a novel generative adversarial network (GAN) for synthesizing large-scale tabular databases which contain various features such as continuous, discrete, and binary. Technically, our GAN belongs to the category of class-conditioned generative models with a predefined conditional vector. However, we propose a new formulation for deriving such a vector incorporating both binary and discrete features simultaneously. We refer to this noble definition as compound conditional vector and employ it for training the generator network. The core architecture of this network is a three-layered deep residual neural network with skip connections. For improving the stability of such complex architecture, we present a regularization scheme towards limiting unprecedented variations on its weight vectors during training. This regularization approach is quite compatible with the nature of adversarial training and it is not computationally prohibitive in runtime. Furthermore, we constantly monitor the variation of the weight vectors for identifying any potential instabilities or irregularities to measure the strength of our proposed regularizer. Toward this end, we also develop a new metric for tracking sudden perturbation on the weight vectors using the singular value decomposition theory. Finally, we evaluate the performance of our proposed synthesis approach on six benchmarking tabular databases, namely Adult, Census, HCDR, Cabs, News, and King. The achieved results corroborate that for the majority of the cases, our proposed RccGAN outperforms other conventional and modern generative models in terms of accuracy, stability, and reliability.

R. Gnana Praveen, Eric Granger, Patrick Cardinal

Multimodal analysis has recently drawn much interest in affective computing, since it can improve the overall accuracy of emotion recognition over isolated uni-modal approaches. The most effective techniques for multimodal emotion recognition efficiently leverage diverse and complimentary sources of information, such as facial, vocal, and physiological modalities, to provide comprehensive feature representations. In this paper, we focus on dimensional emotion recognition based on the fusion of facial and vocal modalities extracted from videos, where complex spatiotemporal relationships may be captured. Most of the existing fusion techniques rely on recurrent networks or conventional attention mechanisms that do not effectively leverage the complimentary nature of audio-visual (A-V) modalities. We introduce a cross-attentional fusion approach to extract the salient features across A-V modalities, allowing for accurate prediction of continuous values of valence and arousal. Our new cross-attentional A-V fusion model efficiently leverages the inter-modal relationships. In particular, it computes cross-attention weights to focus on the more contributive features across individual modalities, and thereby combine contributive feature representations, which are then fed to fully connected layers for the prediction of valence and arousal. The effectiveness of the proposed approach is validated experimentally on videos from the RECOLA and Fatigue (private) data-sets. Results indicate that our cross-attentional A-V fusion model is a cost-effective approach that outperforms state-of-the-art fusion approaches. Code is available: \url{https://github.com/praveena2j/Cross-Attentional-AV-Fusion}

Alexandrine Fortier, Sonal Joshi, Thomas Thebaud et al.

In this work, we propose a multi-target backdoor attack against speaker identification using position-independent clicking sounds as triggers. Unlike previous single-target approaches, our method targets up to 50 speakers simultaneously, achieving success rates of up to 95.04%. To simulate more realistic attack conditions, we vary the signal-to-noise ratio between speech and trigger, demonstrating a trade-off between stealth and effectiveness. We further extend the attack to the speaker verification task by selecting the most similar training speaker - based on cosine similarity - as a proxy target. The attack is most effective when target and enrolled speaker pairs are highly similar, reaching success rates of up to 90% in such cases.

Armita Mohammadi, Alessandro Lameiras Koerich, Laureano Moro-Velazquez et al.

Second language proficiency (L2) in English is usually perceptually evaluated by English teachers or expert evaluators, with the inherent intra- and inter-rater variability. This paper explores deep learning techniques for comprehensive L2 proficiency assessment, addressing both the speech signal and its correspondent transcription. We analyze spoken proficiency classification prediction using diverse architectures, including 2D CNN, frequency-based CNN, ResNet, and a pretrained wav2vec 2.0 model. Additionally, we examine text-based proficiency assessment by fine-tuning a BERT language model within resource constraints. Finally, we tackle the complex task of spontaneous dialogue assessment, managing long-form audio and speaker interactions through separate applications of wav2vec 2.0 and BERT models. Results from experiments on EFCamDat and ANGLISH datasets and a private dataset highlight the potential of deep learning, especially the pretrained wav2vec 2.0 model, for robust automated L2 proficiency evaluation.

Alexandrine Fortier, Thomas Thebaud, Jesús Villalba et al.

Large Language Models (LLMs) and their multimodal extensions are becoming increasingly popular. One common approach to enable multimodality is to cascade domain-specific encoders with an LLM, making the resulting model inherit vulnerabilities from all of its components. In this work, we present the first systematic study of audio backdoor attacks against speech language models. We demonstrate its effectiveness across four speech encoders and three datasets, covering four tasks: automatic speech recognition (ASR), speech emotion recognition, and gender and age prediction. The attack consistently achieves high success rates, ranging from 90.76% to 99.41%. To better understand how backdoors propagate, we conduct a component-wise analysis to identify the most vulnerable stages of the pipeline. Finally, we propose a fine-tuning-based defense that mitigates the threat of poisoned pretrained encoders.

Mohammad Esmaeilpour, Nourhene Chaalia, Adel Abusitta et al.

This paper introduces a bi-discriminator GAN for synthesizing tabular datasets containing continuous, binary, and discrete columns. Our proposed approach employs an adapted preprocessing scheme and a novel conditional term for the generator network to more effectively capture the input sample distributions. Additionally, we implement straightforward yet effective architectures for discriminator networks aiming at providing more discriminative gradient information to the generator. Our experimental results on four benchmarking public datasets corroborates the superior performance of our GAN both in terms of likelihood fitness metric and machine learning efficacy.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

This paper proposes a new defense approach for counteracting state-of-the-art white and black-box adversarial attack algorithms. Our approach fits into the implicit reactive defense algorithm category since it does not directly manipulate the potentially malicious input signals. Instead, it reconstructs a similar signal with a synthesized spectrogram using a cyclic generative adversarial network. This cyclic framework helps to yield a stable generative model. Finally, we feed the reconstructed signal into the speech-to-text model for transcription. The conducted experiments on targeted and non-targeted adversarial attacks developed for attacking DeepSpeech, Kaldi, and Lingvo models demonstrate the proposed defense's effectiveness in adverse scenarios.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

This paper introduces a novel adversarial algorithm for attacking the state-of-the-art speech-to-text systems, namely DeepSpeech, Kaldi, and Lingvo. Our approach is based on developing an extension for the conventional distortion condition of the adversarial optimization formulation using the Cramèr integral probability metric. Minimizing over this metric, which measures the discrepancies between original and adversarial samples' distributions, contributes to crafting signals very close to the subspace of legitimate speech recordings. This helps to yield more robust adversarial signals against playback over-the-air without employing neither costly expectation over transformation operations nor static room impulse response simulations. Our approach outperforms other targeted and non-targeted algorithms in terms of word error rate and sentence-level-accuracy with competitive performance on the crafted adversarial signals' quality. Compared to seven other strong white and black-box adversarial attacks, our proposed approach is considerably more resilient against multiple consecutive playbacks over-the-air, corroborating its higher robustness in noisy environments.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

This paper introduces a defense approach against end-to-end adversarial attacks developed for cutting-edge speech-to-text systems. The proposed defense algorithm has four major steps. First, we represent speech signals with 2D spectrograms using the short-time Fourier transform. Second, we iteratively find a safe vector using a spectrogram subspace projection operation. This operation minimizes the chordal distance adjustment between spectrograms with an additional regularization term. Third, we synthesize a spectrogram with such a safe vector using a novel GAN architecture trained with Sobolev integral probability metric. To improve the model's performance in terms of stability and the total number of learned modes, we impose an additional constraint on the generator network. Finally, we reconstruct the signal from the synthesized spectrogram and the Griffin-Lim phase approximation technique. We evaluate the proposed defense approach against six strong white and black-box adversarial attacks benchmarked on DeepSpeech, Kaldi, and Lingvo models. Our experimental results show that our algorithm outperforms other state-of-the-art defense algorithms both in terms of accuracy and signal quality.

R. Gnana Praveen, Patrick Cardinal, Eric Granger

In the recent years, there has been a shift in facial behavior analysis from the laboratory-controlled conditions to the challenging in-the-wild conditions due to the superior performance of deep learning based approaches for many real world applications.However, the performance of deep learning approaches relies on the amount of training data. One of the major problems with data acquisition is the requirement of annotations for large amount of training data. Labeling process of huge training data demands lot of human support with strong domain expertise for facial expressions or action units, which is difficult to obtain in real-time environments.Moreover, labeling process is highly vulnerable to ambiguity of expressions or action units, especially for intensities due to the bias induced by the domain experts. Therefore, there is an imperative need to address the problem of facial behavior analysis with weak annotations. In this paper, we provide a comprehensive review of weakly supervised learning (WSL) approaches for facial behavior analysis with both categorical as well as dimensional labels along with the challenges and potential research directions associated with it. First, we introduce various types of weak annotations in the context of facial behavior analysis and the corresponding challenges associated with it. We then systematically review the existing state-of-the-art approaches and provide a taxonomy of these approaches along with their insights and limitations. In addition, widely used data-sets in the reviewed literature and the performance of these approaches along with evaluation principles are summarized. Finally, we discuss the remaining challenges and opportunities along with the potential research directions in order to apply facial behavior analysis with weak labels in real life situations.

Gnana Praveen R, Eric Granger, Patrick Cardinal

Automatic estimation of pain intensity from facial expressions in videos has an immense potential in health care applications. However, domain adaptation (DA) is needed to alleviate the problem of domain shifts that typically occurs between video data captured in source and target do-mains. Given the laborious task of collecting and annotating videos, and the subjective bias due to ambiguity among adjacent intensity levels, weakly-supervised learning (WSL)is gaining attention in such applications. Yet, most state-of-the-art WSL models are typically formulated as regression problems, and do not leverage the ordinal relation between intensity levels, nor the temporal coherence of multiple consecutive frames. This paper introduces a new deep learn-ing model for weakly-supervised DA with ordinal regression(WSDA-OR), where videos in target domain have coarse la-bels provided on a periodic basis. The WSDA-OR model enforces ordinal relationships among the intensity levels as-signed to the target sequences, and associates multiple relevant frames to sequence-level labels (instead of a single frame). In particular, it learns discriminant and domain-invariant feature representations by integrating multiple in-stance learning with deep adversarial DA, where soft Gaussian labels are used to efficiently represent the weak ordinal sequence-level labels from the target domain. The proposed approach was validated on the RECOLA video dataset as fully-labeled source domain, and UNBC-McMaster video data as weakly-labeled target domain. We have also validated WSDA-OR on BIOVID and Fatigue (private) datasets for sequence level estimation. Experimental results indicate that our approach can provide a significant improvement over the state-of-the-art models, allowing to achieve a greater localization accuracy.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

In this paper we propose a novel defense approach against end-to-end adversarial attacks developed to fool advanced speech-to-text systems such as DeepSpeech and Lingvo. Unlike conventional defense approaches, the proposed approach does not directly employ low-level transformations such as autoencoding a given input signal aiming at removing potential adversarial perturbation. Instead of that, we find an optimal input vector for a class conditional generative adversarial network through minimizing the relative chordal distance adjustment between a given test input and the generator network. Then, we reconstruct the 1D signal from the synthesized spectrogram and the original phase information derived from the given input signal. Hence, this reconstruction does not add any extra noise to the signal and according to our experimental results, our defense-GAN considerably outperforms conventional defense algorithms both in terms of word error rate and sentence level recognition accuracy.

Mohammad Esmaeilpour, Raymel Alfonso Sallo, Olivier St-Georges et al.

In this paper we propose a conditioning trick, called difference departure from normality, applied on the generator network in response to instability issues during GAN training. We force the generator to get closer to the departure from normality function of real samples computed in the spectral domain of Schur decomposition. This binding makes the generator amenable to truncation and does not limit exploring all the possible modes. We slightly modify the BigGAN architecture incorporating residual network for synthesizing 2D representations of audio signals which enables reconstructing high quality sounds with some preserved phase information. Additionally, the proposed conditional training scenario makes a trade-off between fidelity and variety for the generated spectrograms. The experimental results on UrbanSound8k and ESC-50 environmental sound datasets and the Mozilla common voice dataset have shown that the proposed GAN configuration with the conditioning trick remarkably outperforms baseline architectures, according to three objective metrics: inception score, Frechet inception distance, and signal-to-noise ratio.

Raymel Alfonso Sallo, Mohammad Esmaeilpour, Patrick Cardinal

In this paper, we investigate the potential effect of the adversarially training on the robustness of six advanced deep neural networks against a variety of targeted and non-targeted adversarial attacks. We firstly show that, the ResNet-56 model trained on the 2D representation of the discrete wavelet transform appended with the tonnetz chromagram outperforms other models in terms of recognition accuracy. Then we demonstrate the positive impact of adversarially training on this model as well as other deep architectures against six types of attack algorithms (white and black-box) with the cost of the reduced recognition accuracy and limited adversarial perturbation. We run our experiments on two benchmarking environmental sound datasets and show that without any imposed limitations on the budget allocations for the adversary, the fooling rate of the adversarially trained models can exceed 90\%. In other words, adversarial attacks exist in any scales, but they might require higher adversarial perturbations compared to non-adversarially trained models.

R. Gnana Praveen, Eric Granger, Patrick Cardinal

Estimation of pain intensity from facial expressions captured in videos has an immense potential for health care applications. Given the challenges related to subjective variations of facial expressions, and operational capture conditions, the accuracy of state-of-the-art DL models for recognizing facial expressions may decline. Domain adaptation has been widely explored to alleviate the problem of domain shifts that typically occur between video data captured across various source and target domains. Moreover, given the laborious task of collecting and annotating videos, and subjective bias due to ambiguity among adjacent intensity levels, weakly-supervised learning is gaining attention in such applications. State-of-the-art WSL models are typically formulated as regression problems, and do not leverage the ordinal relationship among pain intensity levels, nor temporal coherence of multiple consecutive frames. This paper introduces a new DL model for weakly-supervised DA with ordinal regression that can be adapted using target domain videos with coarse labels provided on a periodic basis. The WSDA-OR model enforces ordinal relationships among intensity levels assigned to target sequences, and associates multiple relevant frames to sequence-level labels. In particular, it learns discriminant and domain-invariant feature representations by integrating multiple instance learning with deep adversarial DA, where soft Gaussian labels are used to efficiently represent the weak ordinal sequence-level labels from target domain. The proposed approach was validated using RECOLA video dataset as fully-labeled source domain data, and UNBC-McMaster shoulder pain video dataset as weakly-labeled target domain data. We have also validated WSDA-OR on BIOVID and Fatigue datasets for sequence level estimation.

Mohammad Esmaeilpour, Raymel Alfonso Sallo, Olivier St-Georges et al.

In this paper we address the instability issue of generative adversarial network (GAN) by proposing a new similarity metric in unitary space of Schur decomposition for 2D representations of audio and speech signals. We show that encoding departure from normality computed in this vector space into the generator optimization formulation helps to craft more comprehensive spectrograms. We demonstrate the effectiveness of binding this metric for enhancing stability in training with less mode collapse compared to baseline GANs. Experimental results on subsets of UrbanSound8k and Mozilla common voice datasets have shown considerable improvements on the quality of the generated samples measured by the Fréchet inception distance. Moreover, reconstructed signals from these samples, have achieved higher signal to noise ratio compared to regular LS-GANs.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

In this paper, we investigate the impact of different standard environmental sound representations (spectrograms) on the recognition performance and adversarial attack robustness of a victim residual convolutional neural network. Averaged over various experiments on three benchmarking environmental sound datasets, we found the ResNet-18 model outperforms other deep learning architectures such as GoogLeNet and AlexNet both in terms of classification accuracy and the number of training parameters. Therefore we set this model as our front-end classifier for subsequent investigations. Herein, we measure the impact of different settings required for generating more informative mel-frequency cepstral coefficient (MFCC), short-time Fourier transform (STFT), and discrete wavelet transform (DWT) representations on our front-end model. This measurement involves comparing the classification performance over the adversarial robustness. On the balance of average budgets allocated by adversary and the cost of attack, we demonstrate an inverse relationship between recognition accuracy and model robustness against six attack algorithms. Moreover, our experimental results show that while the ResNet-18 model trained on DWT spectrograms achieves the highest recognition accuracy, attacking this model is relatively more costly for the adversary compared to other 2D representations.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

Adversarial attacks have always been a serious threat for any data-driven model. In this paper, we explore subspaces of adversarial examples in unitary vector domain, and we propose a novel detector for defending our models trained for environmental sound classification. We measure chordal distance between legitimate and malicious representation of sounds in unitary space of generalized Schur decomposition and show that their manifolds lie far from each other. Our front-end detector is a regularized logistic regression which discriminates eigenvalues of legitimate and adversarial spectrograms. The experimental results on three benchmarking datasets of environmental sounds represented by spectrograms reveal high detection rate of the proposed detector for eight types of adversarial attacks and outperforms other detection approaches.

R. Gnana Praveen, Eric Granger, Patrick Cardinal

Automatic pain assessment has an important potential diagnostic value for populations that are incapable of articulating their pain experiences. As one of the dominating nonverbal channels for eliciting pain expression events, facial expressions has been widely investigated for estimating the pain intensity of individual. However, using state-of-the-art deep learning (DL) models in real-world pain estimation applications poses several challenges related to the subjective variations of facial expressions, operational capture conditions, and lack of representative training videos with labels. Given the cost of annotating intensity levels for every video frame, we propose a weakly-supervised domain adaptation (WSDA) technique that allows for training 3D CNNs for spatio-temporal pain intensity estimation using weakly labeled videos, where labels are provided on a periodic basis. In particular, WSDA integrates multiple instance learning into an adversarial deep domain adaptation framework to train an Inflated 3D-CNN (I3D) model such that it can accurately estimate pain intensities in the target operational domain. The training process relies on weak target loss, along with domain loss and source loss for domain adaptation of the I3D model. Experimental results obtained using labeled source domain RECOLA videos and weakly-labeled target domain UNBC-McMaster videos indicate that the proposed deep WSDA approach can achieve significantly higher level of sequence (bag)-level and frame (instance)-level pain localization accuracy than related state-of-the-art approaches.

Masih Aminbeidokhti, Marco Pedersoli, Patrick Cardinal et al.

Video-based emotion recognition is a challenging task because it requires to distinguish the small deformations of the human face that represent emotions, while being invariant to stronger visual differences due to different identities. State-of-the-art methods normally use complex deep learning models such as recurrent neural networks (RNNs, LSTMs, GRUs), convolutional neural networks (CNNs, C3D, residual networks) and their combination. In this paper, we propose a simpler approach that combines a CNN pre-trained on a public dataset of facial images with (1) a spatial attention mechanism, to localize the most important regions of the face for a given emotion, and (2) temporal softmax pooling, to select the most important frames of the given video. Results on the challenging EmotiW dataset show that this approach can achieve higher accuracy than more complex approaches.

Sajjad Abdoli, Luiz G. Hafemann, Jerome Rony et al.

We demonstrate the existence of universal adversarial perturbations, which can fool a family of audio classification architectures, for both targeted and untargeted attack scenarios. We propose two methods for finding such perturbations. The first method is based on an iterative, greedy approach that is well-known in computer vision: it aggregates small perturbations to the input so as to push it to the decision boundary. The second method, which is the main contribution of this work, is a novel penalty formulation, which finds targeted and untargeted universal adversarial perturbations. Differently from the greedy approach, the penalty method minimizes an appropriate objective function on a batch of samples. Therefore, it produces more successful attacks when the number of training samples is limited. Moreover, we provide a proof that the proposed penalty method theoretically converges to a solution that corresponds to universal adversarial perturbations. We also demonstrate that it is possible to provide successful attacks using the penalty method when only one sample from the target dataset is available for the attacker. Experimental results on attacking various 1D CNN architectures have shown attack success rates higher than 85.0% and 83.1% for targeted and untargeted attacks, respectively using the proposed penalty method.

Juan D. S. Ortega, Mohammed Senoussaoui, Eric Granger et al.

This paper presents a novel deep neural network (DNN) for multimodal fusion of audio, video and text modalities for emotion recognition. The proposed DNN architecture has independent and shared layers which aim to learn the representation for each modality, as well as the best combined representation to achieve the best prediction. Experimental results on the AVEC Sentiment Analysis in the Wild dataset indicate that the proposed DNN can achieve a higher level of Concordance Correlation Coefficient (CCC) than other state-of-the-art systems that perform early fusion of modalities at feature-level (i.e., concatenation) and late fusion at score-level (i.e., weighted average) fusion. The proposed DNN has achieved CCCs of 0.606, 0.534, and 0.170 on the development partition of the dataset for predicting arousal, valence and liking, respectively.

Mohammed Senoussaoui, Patrick Cardinal, Alessandro Lameiras Koerich

In this paper we present a novel approach for extracting a Bag-of-Words (BoW) representation based on a Neural Network codebook. The conventional BoW model is based on a dictionary (codebook) built from elementary representations which are selected randomly or by using a clustering algorithm on a training dataset. A metric is then used to assign unseen elementary representations to the closest dictionary entries in order to produce a histogram. In the proposed approach, an autoencoder (AE) encompasses the role of both the dictionary creation and the assignment metric. The dimension of the encoded layer of the AE corresponds to the size of the dictionary and the output of its neurons represents the assignment metric. Experimental results for the continuous emotion prediction task on the AVEC 2017 audio dataset have shown an improvement of the Concordance Correlation Coefficient (CCC) from 0.225 to 0.322 for arousal dimension and from 0.244 to 0.368 for valence dimension relative to the conventional BoW version implemented in a baseline system.

Juan D. S. Ortega, Patrick Cardinal, Alessandro L. Koerich

In this paper we propose a fusion approach to continuous emotion recognition that combines visual and auditory modalities in their representation spaces to predict the arousal and valence levels. The proposed approach employs a pre-trained convolution neural network and transfer learning to extract features from video frames that capture the emotional content. For the auditory content, a minimalistic set of parameters such as prosodic, excitation, vocal tract, and spectral descriptors are used as features. The fusion of these two modalities is carried out at a feature level, before training a single support vector regressor (SVR) or at a prediction level, after training one SVR for each modality. The proposed approach also includes preprocessing and post-processing techniques which contribute favorably to improving the concordance correlation coefficient (CCC). Experimental results for predicting spontaneous and natural emotions on the RECOLA dataset have shown that the proposed approach takes advantage of the complementary information of visual and auditory modalities and provides CCCs of 0.749 and 0.565 for arousal and valence, respectively.

Mohammed Senoussaoui, Patrick Cardinal, Najim Dehak et al.

Automatic measuring of speaker sincerity degree is a novel research problem in computational paralinguistics. This paper proposes covariance-based feature vectors to model speech and ensembles of support vector regressors to estimate the degree of sincerity of a speaker. The elements of each covariance vector are pairwise statistics between the short-term feature components. These features are used alone as well as in combination with the ComParE acoustic feature set. The experimental results on the development set of the Sincerity Speech Corpus using a cross-validation procedure have shown an 8.1% relative improvement in the Spearman's correlation coefficient over the baseline system.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

Adversarial audio attacks can be considered as a small perturbation unperceptive to human ears that is intentionally added to the audio signal and causes a machine learning model to make mistakes. This poses a security concern about the safety of machine learning models since the adversarial attacks can fool such models toward the wrong predictions. In this paper we first review some strong adversarial attacks that may affect both audio signals and their 2D representations and evaluate the resiliency of the most common machine learning model, namely deep learning models and support vector machines (SVM) trained on 2D audio representations such as short time Fourier transform (STFT), discrete wavelet transform (DWT) and cross recurrent plot (CRP) against several state-of-the-art adversarial attacks. Next, we propose a novel approach based on pre-processed DWT representation of audio signals and SVM to secure audio systems against adversarial attacks. The proposed architecture has several preprocessing modules for generating and enhancing spectrograms including dimension reduction and smoothing. We extract features from small patches of the spectrograms using speeded up robust feature (SURF) algorithm which are further used to generate a codebook using the K-Means++ algorithm. Finally, codewords are used to train a SVM on the codebook of the SURF-generated vectors. All these steps yield to a novel approach for audio classification that provides a good trade-off between accuracy and resilience. Experimental results on three environmental sound datasets show the competitive performance of proposed approach compared to the deep neural networks both in terms of accuracy and robustness against strong adversarial attacks.

Sajjad Abdoli, Patrick Cardinal, Alessandro Lameiras Koerich

In this paper, we present an end-to-end approach for environmental sound classification based on a 1D Convolution Neural Network (CNN) that learns a representation directly from the audio signal. Several convolutional layers are used to capture the signal's fine time structure and learn diverse filters that are relevant to the classification task. The proposed approach can deal with audio signals of any length as it splits the signal into overlapped frames using a sliding window. Different architectures considering several input sizes are evaluated, including the initialization of the first convolutional layer with a Gammatone filterbank that models the human auditory filter response in the cochlea. The performance of the proposed end-to-end approach in classifying environmental sounds was assessed on the UrbanSound8k dataset and the experimental results have shown that it achieves 89% of mean accuracy. Therefore, the propose approach outperforms most of the state-of-the-art approaches that use handcrafted features or 2D representations as input. Furthermore, the proposed approach has a small number of parameters compared to other architectures found in the literature, which reduces the amount of data required for training.

Mohammad Esmaeilpour, Patrick Cardinal, Alessandro Lameiras Koerich

In this paper we propose a novel environmental sound classification approach incorporating unsupervised feature learning from codebook via spherical $K$-Means++ algorithm and a new architecture for high-level data augmentation. The audio signal is transformed into a 2D representation using a discrete wavelet transform (DWT). The DWT spectrograms are then augmented by a novel architecture for cycle-consistent generative adversarial network. This high-level augmentation bootstraps generated spectrograms in both intra and inter class manners by translating structural features from sample to sample. A codebook is built by coding the DWT spectrograms with the speeded-up robust feature detector (SURF) and the K-Means++ algorithm. The Random Forest is our final learning algorithm which learns the environmental sound classification task from the clustered codewords in the codebook. Experimental results in four benchmarking environmental sound datasets (ESC-10, ESC-50, UrbanSound8k, and DCASE-2017) have shown that the proposed classification approach outperforms the state-of-the-art classifiers in the scope, including advanced and dense convolutional neural networks such as AlexNet and GoogLeNet, improving the classification rate between 3.51% and 14.34%, depending on the dataset.

Ahmed Ali, Najim Dehak, Patrick Cardinal et al.

We investigate different approaches for dialect identification in Arabic broadcast speech, using phonetic, lexical features obtained from a speech recognition system, and acoustic features using the i-vector framework. We studied both generative and discriminate classifiers, and we combined these features using a multi-class Support Vector Machine (SVM). We validated our results on an Arabic/English language identification task, with an accuracy of 100%. We used these features in a binary classifier to discriminate between Modern Standard Arabic (MSA) and Dialectal Arabic, with an accuracy of 100%. We further report results using the proposed method to discriminate between the five most widely used dialects of Arabic: namely Egyptian, Gulf, Levantine, North African, and MSA, with an accuracy of 52%. We discuss dialect identification errors in the context of dialect code-switching between Dialectal Arabic and MSA, and compare the error pattern between manually labeled data, and the output from our classifier. We also release the train and test data as standard corpus for dialect identification.