Speech Enhancement based on cascaded two flows
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_ | 1866916907490213888 |
|---|---|
| author | Lee, Seonggyu Cheong, Sein Han, Sangwook Kim, Kihyuk Shin, Jong Won |
| author_facet | Lee, Seonggyu Cheong, Sein Han, Sangwook Kim, Kihyuk Shin, Jong Won |
| contents | Speech enhancement (SE) based on diffusion probabilistic models has exhibited impressive performance, while requiring a relatively high number of function evaluations (NFE). Recently, SE based on flow matching has been proposed, which showed competitive performance with a small NFE. Early approaches adopted the noisy speech as the only conditioning variable. There have been other approaches which utilize speech enhanced with a predictive model as another conditioning variable and to sample an initial value, but they require a separate predictive model on top of the generative SE model. In this work, we propose to employ an identical model based on flow matching for both SE and generating enhanced speech used as an initial starting point and a conditioning variable. Experimental results showed that the proposed method required the same or fewer NFEs even with two cascaded generative methods while achieving equivalent or better performances to the previous baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_06842 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Speech Enhancement based on cascaded two flows Lee, Seonggyu Cheong, Sein Han, Sangwook Kim, Kihyuk Shin, Jong Won Audio and Speech Processing Signal Processing Speech enhancement (SE) based on diffusion probabilistic models has exhibited impressive performance, while requiring a relatively high number of function evaluations (NFE). Recently, SE based on flow matching has been proposed, which showed competitive performance with a small NFE. Early approaches adopted the noisy speech as the only conditioning variable. There have been other approaches which utilize speech enhanced with a predictive model as another conditioning variable and to sample an initial value, but they require a separate predictive model on top of the generative SE model. In this work, we propose to employ an identical model based on flow matching for both SE and generating enhanced speech used as an initial starting point and a conditioning variable. Experimental results showed that the proposed method required the same or fewer NFEs even with two cascaded generative methods while achieving equivalent or better performances to the previous baselines. |
| title | Speech Enhancement based on cascaded two flows |
| topic | Audio and Speech Processing Signal Processing |
| url | https://arxiv.org/abs/2508.06842 |