arXiv subjects

Jing-Tong Tzeng

Publications and source records attributed to Jing-Tong Tzeng.

3 recordsLinked to original sources

arXiv2026

Joint Learning using Mixture-of-Expert-Based Representation for Speech Enhancement and Robust Emotion Recognition

Speech emotion recognition (SER) plays a critical role in building emotion-aware speech systems, but its performance degrades significantly under noisy conditions. Although speech enhancement (SE) can improve robustness, it often introduces artifacts that obscure emotional cues and adds computational overhead to the pipeline. Multi-task learning (MTL) offers an alternative by jointly optimizing SE and SER tasks. However, conventional shared-backbone models frequently suffer from gradient interference and representational conflicts between tasks. To address these challenges, we propose the Sparse Mixture-of-Experts Representation Integration Technique (Sparse MERIT), a flexible MTL framework that applies frame-wise expert routing over self-supervised speech representations. Sparse MERIT incorporates task-specific gating networks that dynamically select from a shared pool of experts for each frame, enabling parameter-efficient and task-adaptive representation learning. Experiments on the MSP-Podcast corpus show that Sparse MERIT consistently outperforms baseline models on both SER and SE tasks. Under the most challenging condition of -5 dB signal-to-noise ratio (SNR), Sparse MERIT improves SER F1-macro by an average of 12.0% over a baseline relying on a SE pre-processing strategy, and by 3.4% over a naive MTL baseline, with statistical significance on unseen noise conditions. For SE, Sparse MERIT improves segmental SNR (SSNR) by 28.2% over the SE pre-processing baseline and by 20.0% over the naive MTL baseline. These results demonstrate that Sparse MERIT provides robust and generalizable performance for both emotion recognition and enhancement tasks in noisy environments.

eess.AS↗

arXiv2025

Lessons Learnt: Revisit Key Training Strategies for Effective Speech Emotion Recognition in the Wild

In this study, we revisit key training strategies in machine learning often overlooked in favor of deeper architectures. Specifically, we explore balancing strategies, activation functions, and fine-tuning techniques to enhance speech emotion recognition (SER) in naturalistic conditions. Our findings show that simple modifications improve generalization with minimal architectural changes. Our multi-modal fusion model, integrating these optimizations, achieves a valence CCC of 0.6953, the best valence score in Task 2: Emotional Attribute Regression. Notably, fine-tuning RoBERTa and WavLM separately in a single-modality setting, followed by feature fusion without training the backbone extractor, yields the highest valence performance. Additionally, focal loss and activation functions significantly enhance performance without increasing complexity. These results suggest that refining core components, rather than deepening models, leads to more robust SER in-the-wild.

eess.AS↗

arXiv2025

Improving the Robustness and Clinical Applicability of Automatic Respiratory Sound Classification Using Deep Learning-Based Audio Enhancement: Algorithm Development and Validation

Deep learning techniques have shown promising results in the automatic classification of respiratory sounds. However, accurately distinguishing these sounds in real-world noisy conditions remains challenging for clinical deployment. In addition, predicting signals with only background noise may reduce user trust in the system. This study explores the feasibility and effectiveness of incorporating a deep learning-based audio enhancement step into automatic respiratory sound classification systems to improve robustness and clinical applicability. We conducted extensive experiments using various audio enhancement model architectures, including time-domain and time-frequency-domain approaches, combined with multiple classification models to evaluate the module's effectiveness. The classification performance was compared against the noise injection data augmentation method. These experiments were carried out on two datasets: the ICBHI respiratory sound dataset and the FABS dataset. Furthermore, a physician validation study assessed the system's clinical utility. Integrating the audio enhancement module resulted in a 21.9% increase in the ICBHI classification score and a 4.1% improvement on the FABS dataset in multi-class noisy scenarios. Quantitative analysis revealed efficiency gains, higher diagnostic confidence, and increased trust, with workflows using enhanced audio improving diagnostic sensitivity by 11.6% and enabling high-confidence diagnoses. Incorporating an audio enhancement algorithm boosts the robustness and clinical utility of automatic respiratory sound classification systems, enhancing performance in noisy environments and fostering greater trust among medical professionals.

eess.AS↗