Arabic speech emotion recognition (SER) in low-resource corpora demands comprehensive evaluation for trustworthy and reproducible outcomes. In this work, we rigorously analyze the EYASE corpus through six controlled experiments using handcrafted features, pretrained representations, and fine-tuned speech models. The proposed hybrid pipeline combines wav2vec2 self-supervised embeddings, handcrafted Spectro-temporal statistics and pitch descriptors via a soft-voting extra trees XGBoost ensemble. It achieves 72.4% accuracy and 0.72 macro-F1 on a held-out test split and 70.17 ± 1.78% accuracy over five seeds, surpassing the alternatives we analyze, including a CNN-BiLSTM baseline (60.3%) and fine-tuned emotion2vec (64.8 ± 3.3%). Feature-block ablation exposes high redundancy, where no block is needed. SHAP, LIME and Grad-CAM analyses reveal factors influencing forecasts. Finally, statistical confirmation of the observed performance differences is provided by Welch’s t test with Holm correction. Overall, this work offers a reproducible and interpretable standard for low-resource Egyptian Arabic SER and provides insights into various speech representations.
A systematic evaluation of acoustic representation and feature fusion strategies for Arabic speech emotion recognition. // Life Cycle Reliability and Safety Engineering. – 2026. – DOI: 10.1007/s10772-026-10290-9. – URL: https://link.springer.com/article/10.1007/s10772-026-10290-9
(2026). A systematic evaluation of acoustic representation and feature fusion strategies for Arabic speech emotion recognition. Life Cycle Reliability and Safety Engineering. https://doi.org/10.1007/s10772-026-10290-9
“A systematic evaluation of acoustic representation and feature fusion strategies for Arabic speech emotion recognition.” Life Cycle Reliability and Safety Engineering, 2026. doi:10.1007/s10772-026-10290-9.
@article{article35623,
title = {A systematic evaluation of acoustic representation and feature fusion strategies for Arabic speech emotion recognition},
journal = {Life Cycle Reliability and Safety Engineering},
year = {2026},
doi = {10.1007/s10772-026-10290-9},
url = {https://link.springer.com/article/10.1007/s10772-026-10290-9}
}
Phonetic similarity between commands can be a source of recognition errors in small-vocabulary speech recognition systems. This study presents an embedded automatic speech recognition system for 35 Modern Standard Arabic industrial...
Wideband (WB) speech, spanning 0–8 kHz, is not transmitted over Public Switch Telephone Networks (PSTNs) due to spectrum limitations. Instead, only Restricted Band (RB) speech up to 4 kHz is supported, which severely degrades speech...
Studies on dialect identification have increased, with majority leveraging simple to complex classifiers. However, these complex systems exhibit opaque characteristics, such that their decision-making process is not transparent to the...
The “Deep multimodal K-fold model” is an approach to machine learning that uses multiple modalities, such as text and audio, to analyze emotion and sentiment in Figurative Language. This model uses a K-fold cross-validation method to...
Automatic Speech Recognition (ASR) systems trained on native English accents often underperform for Indian-accented English due to distinct phonetic and prosodic variations. This study presents a comprehensive, architecture-controlled...