Metraipolis › archive › dc_fc4002b8b9af
PASE: Leveraging the Phonological Prior of WavLM for Low-Hallucination Generative Speech Enhancement — arXiv:2511.13300
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2511.13300
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:17:21.603Z
- SHA-256 of the archived text
7f714aadfd2e3dc2941f9ea485237980a5374b231601347c68fcb508be0fae74- Stable URL of this page
https://metraipolis.com/documento/dc_fc4002b8b9af
How to cite
PASE: Leveraging the Phonological Prior of WavLM for Low-Hallucination Generative Speech Enhancement — arXiv:2511.13300 — arXiv (en), https://arxiv.org/abs/2511.13300 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_fc4002b8b9af (Metraipolis archive dc_fc4002b8b9af, sha256 7f714aadfd2e3dc2941f9ea485237980a5374b231601347c68fcb508be0fae74)
BibTeX
@misc{dcfc4002b8b9af,
title = {PASE: Leveraging the Phonological Prior of WavLM for Low-Hallucination Generative Speech Enhancement — arXiv:2511.13300},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2511.13300},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_fc4002b8b9af; sha256 7f714aadfd2e3dc2941f9ea485237980a5374b231601347c68fcb508be0fae74},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2511.13300. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Generative models have shown remarkable performance in speech enhancement (SE), achieving superior perceptual quality over traditional discriminative approaches. However, existing generative SE approaches often overlook the risk of hallucination under severe noise, leading to incorrect spoken content or inconsistent speaker characteristics, which we term linguistic and acoustic hallucinations, respectively. We argue that linguistic hallucination stems from models' failure to constrain valid phonological structures and it is a more fundamental challenge. While language models (LMs) are well-suited for capturing the underlying speech structure through modeling the distribution of discrete tokens, existing approaches are limited in learning from noise-corrupted representations, which can lead to contaminated priors and hallucinations. To overcome these limitations, we propose the Phonologically Anchored Speech Enhancer (PASE), a generative SE framework that leverages the robust phonological prior embedded in the pre-trained WavLM model to mitigate hallucinations. First, we adapt WavLM into a denoising expert via representation distillation to clean its final-layer features. Guided by the model's intrinsic phonological prior, this process enables robust denoising while minimizing linguistic hallucinations. To further reduce acoustic hallucinations, we train the vocoder with a dual-stream representation: the high-level phonetic representation provides clean linguistic content, while a low-level acoustic representation retains speaker identity and prosody. Experimental results demonstrate that PASE not only surpasses state-of-the-art discriminative models in perceptual quality, but also significantly outperforms prior generative models with substantially lower linguistic and acoustic hallucinations. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.AI). Categorie arXiv: Audio and Speech Processing (eess.AS) ; Artificial Intelligence (cs.AI); Sound (cs.SD). Pagina del lavoro: https://arxiv.org/abs/2511.13300 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-31T12:05:52.880Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).