Alaa Nfissi

Mr. Alaa Nfissi

CIISE, Concordia University, Montréal, Quebec, Canada
orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
31 May 2024 in Applied Intelligence
Verified
Journal Article
Closed access
PDF file
Alaa NfissiAlaa Nfissi
Alaa Nfissi
WBWassim Bouachir
Wassim Bouachir
Nizar BouguilaNizar Bouguila
Nizar Bouguila
BMBrian Mishara
Brian Mishara

1 December 2022
Verified
Conference Paper
Closed access
Alaa NfissiAlaa Nfissi
Alaa Nfissi
WBWassim Bouachir
Wassim Bouachir
NBNizar Bouguila
Nizar Bouguila
BMBrian L. Mishara
Brian L. Mishara
We present CNN-n-GRU, a new end-to-end (E2E) architecture built of an n-layer convolutional neural network (CNN) followed sequentially by an n-layer Gated Recurrent Unit (GRU) for speech emotion recognition. CNNs and RNNs both exhibited promising outcomes when fed raw waveform voice inputs. This inspired our idea to combine them into a single model to maximise their potential. Instead of using handcrafted features or spectrograms, we train CNNs to recognise low-level speech representations from raw waveform, which allows the network to capture relevant narrow-band emotion characteristics. On the other hand, RNNs (GRUs in our case) can learn temporal characteristics, allowing the network to better capture the signal’s time-distributed features. Because a CNN can generate multiple levels of representation abstraction, we exploit early layers to extract high-level features, then to supply the appropriate input to subsequent RNN layers in order to aggregate long-term dependencies. By taking advantage of both CNNs and GRUs in a single model, the proposed architecture has important advantages over other models from the literature. The proposed model was evaluated using the TESS dataset and compared to state-of-the-art methods. Our experimental results demonstrate that the proposed model is more accurate than traditional classification approaches for speech emotion recognition.

13 February 2025 in IEEE Transactions on Affective Computing
Verified
Journal Article
Closed access
Alaa NfissiAlaa Nfissi
Alaa Nfissi
WBWassim Bouachir
Wassim Bouachir
Nizar BouguilaNizar Bouguila
Nizar Bouguila
BMBrian Mishara
Brian Mishara
In the field of human-computer interaction and psychological assessment, speech emotion recognition (SER) plays an important role in deciphering emotional states from speech signals. Despite advancements, challenges persist due to system complexity, feature distinctiveness issues, and noise interference. This paper introduces a new end-to-end (E2E) deep learning multi-resolution framework for SER, addressing these limitations by extracting meaningful representations directly from raw waveform speech signals. By leveraging the properties of the fast discrete wavelet transform (FDWT), including the cascade algorithm, conjugate quadrature filter, and coefficient denoising, our approach introduces a learnable model for both wavelet bases and denoising through deep learning techniques. The framework incorporates an activation function for learnable asymmetric hard thresholding of wavelet coefficients. Our approach exploits the capabilities of wavelets for effective localization in both time and frequency domains. We then combine one-dimensional dilated convolutional neural networks (1D dilated CNN) with a spatial attention layer and bidirectional gated recurrent units (Bi-GRU) with a temporal attention layer to efficiently capture the nuanced spatial and temporal characteristics of emotional features. By handling variable-length speech without segmentation and eliminating the need for pre or post-processing, the proposed model outperformed state-of-the-art methods on IEMOCAP and EMO-DB datasets. The source code of this paper is shared on the Github repository: https://github.com/alaaNfissi/SigWavNet-Learning-Multiresolution-Signal-Wavelet-Network-for-Speech-Emotion-Recognition.

5 February 2024
Verified
Conference Paper
Closed access
Alaa NfissiAlaa Nfissi
Alaa Nfissi
WBWassim Bouachir
Wassim Bouachir
NBNizar Bouguila
Nizar Bouguila
BMBrian Mishara
Brian Mishara
Suicide remains a major public health concern worldwide, and early detection of suicidal ideation is crucial for prevention. One promising approach for monitoring symptoms is through the prediction of suicidal speech, as speech can be passively collected and may provide insight into changes in risk. However, identifying suicidal speech is challenging due to the rapid variability in speech characteristics and the association of suicidal ideation with emotion dysregulation. In light of these challenges, we present a novel end-to-end (E2E) method for speech emotion recognition (SER) as a mean of detecting changes in emotional state, that may indicate a high risk of suicide. Our method incorporates the use of Convolutional Neural Networks (CNNs) and Gated Recurrent Units (GRUs) to analyze raw waveform signals. Firstly, our model is designed to train the CNN component to identify higher-level speech representations directly from raw waveform data, rather than relying on manually crafted features or spectrograms. This enables the network to effectively capture specific emotion-related features within a narrow frequency range, while also handling speech of varying lengths without the need for segmentation. Secondly, the GRU component is capable of learning temporal patterns, enhancing the network’s ability to capture time-dependent features in signal. Furthermore, we validate our approach on the NSPLCRISE emotion dataset that we recently created. This dataset contains phone call recordings from lifeline frequent callers with psychological problems, and potentially with a history of suicidal ideation and previous attempts. Our experimental results show that our method outperforms other state-of-the-art techniques for SER. The source code of this paper is provided on the Git repository Unlocking the Emotional States of High-Risk Suicide Callers through Speech Analysis.


15 December 2023
Verified
Conference Paper
Open access
Alaa NfissiAlaa Nfissi
Alaa Nfissi
WBWassim Bouachir
Wassim Bouachir
Nizar BouguilaNizar Bouguila
Nizar Bouguila
BMBrian L. Mishara
Brian L. Mishara
In speech emotion recognition (SER), using predefined features without considering their practical importance may lead to high dimensional datasets, including redundant and irrelevant information. Consequently, high-dimensional learning often results in decreasing model accuracy while increasing computational complexity. Our work underlines the importance of carefully considering and analyzing features in order to build efficient SER systems. We present a new supervised SER method based on an efficient feature engineering approach. We pay particular attention to the explainability of results to evaluate feature relevance and refine feature sets. This is performed iteratively through feature evaluation loop, using Shapley values to boost feature selection and improve overall framework performance. Our approach allows thus to balance the benefits between model performance and transparency. The proposed method outperforms human-level performance (HLP) and state-of-the-art machine learning methods in emotion recognition on the TESS dataset. The source code of this paper is publicly available at Iterative-Feature-Boosting-for-Explainable-Speech-Emotion-Recoanition.