Diffusion-based Speech Enhancement with Schrödinger Bridge and Symmetric Noise Schedule
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916392537686016 |
|---|---|
| author | Wang, Siyi Liu, Siyi Harper, Andrew Kendrick, Paul Salzmann, Mathieu Cernak, Milos |
| author_facet | Wang, Siyi Liu, Siyi Harper, Andrew Kendrick, Paul Salzmann, Mathieu Cernak, Milos |
| contents | Recently, diffusion-based generative models have demonstrated remarkable performance in speech enhancement tasks. However, these methods still encounter challenges, including the lack of structural information and poor performance in low Signal-to-Noise Ratio (SNR) scenarios. To overcome these challenges, we propose the Schröodinger Bridge-based Speech Enhancement (SBSE) method, which learns the diffusion processes directly between the noisy input and the clean distribution, unlike conventional diffusion-based speech enhancement systems that learn data to Gaussian distributions. To enhance performance in extremely noisy conditions, we introduce a two-stage system incorporating ratio mask information into the diffusion-based generative model. Our experimental results show that our proposed SBSE method outperforms all the baseline models and achieves state-of-the-art performance, especially in low SNR conditions. Importantly, only a few inference steps are required to achieve the best result. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_05116 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Diffusion-based Speech Enhancement with Schrödinger Bridge and Symmetric Noise Schedule Wang, Siyi Liu, Siyi Harper, Andrew Kendrick, Paul Salzmann, Mathieu Cernak, Milos Audio and Speech Processing Sound Recently, diffusion-based generative models have demonstrated remarkable performance in speech enhancement tasks. However, these methods still encounter challenges, including the lack of structural information and poor performance in low Signal-to-Noise Ratio (SNR) scenarios. To overcome these challenges, we propose the Schröodinger Bridge-based Speech Enhancement (SBSE) method, which learns the diffusion processes directly between the noisy input and the clean distribution, unlike conventional diffusion-based speech enhancement systems that learn data to Gaussian distributions. To enhance performance in extremely noisy conditions, we introduce a two-stage system incorporating ratio mask information into the diffusion-based generative model. Our experimental results show that our proposed SBSE method outperforms all the baseline models and achieves state-of-the-art performance, especially in low SNR conditions. Importantly, only a few inference steps are required to achieve the best result. |
| title | Diffusion-based Speech Enhancement with Schrödinger Bridge and Symmetric Noise Schedule |
| topic | Audio and Speech Processing Sound |
| url | https://arxiv.org/abs/2409.05116 |