A Comparative Evaluation of Deep Learning Models for Speech Enhancement in Real-World Noisy Environments

TL;DR

This study benchmarks three deep learning models for speech enhancement in noisy environments, evaluating their noise suppression, perceptual quality, and speaker feature retention across multiple datasets.

Contribution

It provides a comprehensive comparative analysis of Wave-U-Net, CMGAN, and U-Net, highlighting their strengths and trade-offs in real-world speech enhancement tasks.

Findings

U-Net achieves highest noise suppression with significant SNR improvements.

CMGAN attains the best perceptual quality with top PESQ scores.

Wave-U-Net balances noise suppression and speaker feature retention.

Abstract

Speech enhancement, particularly denoising, is vital in improving the intelligibility and quality of speech signals for real-world applications, especially in noisy environments. While prior research has introduced various deep learning models for this purpose, many struggle to balance noise suppression, perceptual quality, and speaker-specific feature preservation, leaving a critical research gap in their comparative performance evaluation. This study benchmarks three state-of-the-art models Wave-U-Net, CMGAN, and U-Net, on diverse datasets such as SpEAR, VPQAD, and Clarkson datasets. These models were chosen due to their relevance in the literature and code accessibility. The evaluation reveals that U-Net achieves high noise suppression with SNR improvements of +71.96% on SpEAR, +64.83% on VPQAD, and +364.2% on the Clarkson dataset. CMGAN outperforms in perceptual quality, attaining the highest PESQ scores of 4.04 on SpEAR and 1.46 on VPQAD, making it well-suited for applications prioritizing natural and intelligible speech. Wave-U-Net balances these attributes with improvements in speaker-specific feature retention, evidenced by VeriSpeak score gains of +10.84% on SpEAR and +27.38% on VPQAD. This research indicates how advanced methods can optimize trade-offs between noise suppression, perceptual quality, and speaker recognition. The findings may contribute to advancing voice biometrics, forensic audio analysis, telecommunication, and speaker verification in challenging acoustic conditions.