High-Fidelity Speech Enhancement via Discrete Audio Tokens
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909821534470144 |
|---|---|
| author | Lanzendörfer, Luca A. Berdoz, Frédéric Asonitis, Antonis Wattenhofer, Roger |
| author_facet | Lanzendörfer, Luca A. Berdoz, Frédéric Asonitis, Antonis Wattenhofer, Roger |
| contents | Recent autoregressive transformer-based speech enhancement (SE) methods have shown promising results by leveraging advanced semantic understanding and contextual modeling of speech. However, these approaches often rely on complex multi-stage pipelines and low sampling rate codecs, limiting them to narrow and task-specific speech enhancement. In this work, we introduce DAC-SE1, a simplified language model-based SE framework leveraging discrete high-resolution audio representations; DAC-SE1 preserves fine-grained acoustic details while maintaining semantic coherence. Our experiments show that DAC-SE1 surpasses state-of-the-art autoregressive SE methods on both objective perceptual metrics and in a MUSHRA human evaluation. We release our codebase and model checkpoints to support further research in scalable, unified, and high-quality speech enhancement. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_02187 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | High-Fidelity Speech Enhancement via Discrete Audio Tokens Lanzendörfer, Luca A. Berdoz, Frédéric Asonitis, Antonis Wattenhofer, Roger Sound Machine Learning Audio and Speech Processing Recent autoregressive transformer-based speech enhancement (SE) methods have shown promising results by leveraging advanced semantic understanding and contextual modeling of speech. However, these approaches often rely on complex multi-stage pipelines and low sampling rate codecs, limiting them to narrow and task-specific speech enhancement. In this work, we introduce DAC-SE1, a simplified language model-based SE framework leveraging discrete high-resolution audio representations; DAC-SE1 preserves fine-grained acoustic details while maintaining semantic coherence. Our experiments show that DAC-SE1 surpasses state-of-the-art autoregressive SE methods on both objective perceptual metrics and in a MUSHRA human evaluation. We release our codebase and model checkpoints to support further research in scalable, unified, and high-quality speech enhancement. |
| title | High-Fidelity Speech Enhancement via Discrete Audio Tokens |
| topic | Sound Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2510.02187 |