arXiv · 2611.22628
Back to Mel: Learning to Bridge Fidelity and Perception in Generative Speech Enhancement
Abstract
Recent generative techniques have facilitated the rapid proliferation of speech enhancement approaches built upon learnable compact latent representations and large generative backbones. Nevertheless, it remains inherently challenging to balance between acoustic fidelity, perceptual quality, and processing efficiency. In this paper, we revisit the Mel spectrum as a structured acoustic representation and propose BAME, a flow-matching-based framework that exploits the complementary priors for generative speech enhancement task. First, leveraging the explicit hierarchical frequency distribution of Mel spectrum, we elaborately devise a subband-aware Transformer (SaT) for efficient generative modeling. Using only 2.9% of the parameters, the proposed SaT achieves superior enhancement performance compared with a 331 M DiT backbone. Second, motivated by range-null decomposition theory of Mel filtering in the magnitude-spectral domain, we introduce a phase-aware neural vocoder that incorporates degraded-phase prior, together with a quality-guided dual-phase strategy to adaptively mitigate the effects of unreliable phase information. Third, we devise a simple yet effective one-step fidelity-perception alignment strategy that fine-tunes the flow model with complementary constraints from fidelity and perception domains. Extensive experiments are conducted on DNS-Challenge, EARS, GSR, and URGENT benchmarks, and both objective and subjective results demonstrate promising performance on various intrusive and non-intrusive evaluation metrics. These results further advocate the potential of exploiting structured acoustic representations for reliable and efficient generative speech enhancement. Audio samples and codes are available at https://github.com/AndongLi-speech/BAME.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Andong Li, Kun Zhou, Xiaodong Li, Chengshi Zheng. 2026-11-17. Back to Mel: Learning to Bridge Fidelity and Perception in Generative Speech Enhancement. https://searcharxiv.com/abs/2611.22628
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.