aiqbench analysis
This study addresses the underexplored challenge of Arabic Speech Emotion Recognition by systematically comparing hybrid deep learning architectures. The research demonstrates that a CNN-Transformer model, which combines local feature extraction with global context modeling, significantly outperforms both CNN-LSTM and fine-tuned wav2vec 2.0 approaches on key datasets. For practitioners in low-resource language domains, the findings suggest that integrating Transformer-based context modeling into hybrid pipelines is a viable strategy for improving accuracy in dialectally diverse scenarios.
Towards Robust Arabic Speech Emotion Recognition with Deep Learning
arXiv:2606.10278v1 Announce Type: cross Abstract: Speech Emotion Recognition (SER) aims to identify a speaker's emotional state from audio signals. While recent advances in deep learning have significantly improved SER performance in Indo-European languages, Arabic SER remains underexplored and challenging due to dialectal diversity, limited annotated datasets, and the difficulty of modeling both local spectral cues and long-range temporal dependencies. To address these limitations, this study investigates whether hybrid architectures that jointly model spatial and contextual information can improve emotion recognition in Arabic speech. We propose and evaluate a comparative framework involving three architectures: a CNN-LSTM model, a CNN-Transformer model, and a fine-tuned wav2vec 2.0 model. The first two models leverage MFCC and spectrogram-based representations, while wav2vec 2.0 operates directly on raw audio through self-supervised representations. Experiments conducted on the EYASE and BAVED datasets demonstrate that the proposed CNN-Transformer architecture significantly outperforms the other models, achieving an accuracy of 98.1 percent. This result highlights the effectiveness of combining convolutional feature extraction with Transformer-based global context modeling. The main contribution of this work lies in providing a systematic comparison of hybrid and self-supervised approaches for Arabic SER, and in demonstrating that CNN-Transformer architectures offer a robust solution for capturing both spectral and long-range dependencies in low-resource and dialectally diverse settings.
Abstract:Speech Emotion Recognition (SER) aims to identify a speaker's emotional state from audio signals. While recent advances in deep learning have significantly improved SER performance in Indo-European languages, Arabic SER remains underexplored and challenging due to dialectal diversity, limited annotated datasets, and the difficulty of modeling both local spectral cues and long-range temporal dependencies. To address these limitations, this study investigates whether hybrid architectures that jointly model spatial and contextual information can improve emotion recognition in Arabic speech. We propose and evaluate a comparative framework involving three architectures: a CNN-LSTM model, a CNN-Transformer model, and a fine-tuned wav2vec 2.0 model. The first two models leverage MFCC and spectrogram-based representations, while wav2vec 2.0 operates directly on raw audio through self-supervised representations. Experiments conducted on the EYASE and BAVED datasets demonstrate that the proposed CNN-Transformer architecture significantly outperforms the other models, achieving an accuracy of 98.1 percent. This result highlights the effectiveness of combining convolutional feature extraction with Transformer-based global context modeling. The main contribution of this work lies in providing a systematic comparison of hybrid and self-supervised approaches for Arabic SER, and in demonstrating that CNN-Transformer architectures offer a robust solution for capturing both spectral and long-range dependencies in low-resource and dialectally diverse settings.
21 pages, 16 figures, 11 tables. Submitted manuscript
Sound (cs.SD); Artificial Intelligence (cs.AI)
arXiv:2606.10278 [cs.SD]
(or arXiv:2606.10278v1 [cs.SD] for this version)
https://doi.org/10.48550/arXiv.2606.10278 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 307 words
Related coverage
BRICS Draws a New Global Line: Full Text of New Delhi Declaration
slguardian.org · Sep 12, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026
Microsoft’s AI chief says superintelligence is near, but won’t take your job
The Verge · Jun 8, 2026
Enhancing Multimodal Emotion Recognition via Multi-Feature Encoding and Attention-Based Fusion
arXiv · Sep 7, 2026