Title | ||
---|---|---|
SERIL: Noise Adaptive Speech Enhancement using Regularization-based Incremental Learning |
Abstract | ||
---|---|---|
Numerous noise adaptation techniques have been proposed to address the mismatch problem in speech enhancement (SE) by fine-tuning deep-learning (DL)-based models. However, adaptation to a target domain can lead to catastrophic forgetting of the previously learnt noise environments. Because SE models are commonly used in embedded devices, re-visiting previous noise environments is a common situation in speech enhancement. In this paper, we propose a novel regularization-based incremental learning SE (SERIL) strategy, which can complement these noise adaptation strategies without having to access previous training data. The experimental results show that, when faced with a new noise domain, the SERIL model outperforms the unadapted SE model in various metrics: PESQ, STOI, eSTOI, and short-time spectral amplitude SDR. Meanwhile, compared with the traditional fine-tuning adaptive SE model, the SERIL model can significantly reduce the forgetting of previous noise environments by 52%. The promising results indicate that the SERIL model can effectively overcome the catastrophic forgetting problem and can be suitably deployed in real-world applications, where the noise environment changes frequently. |
Year | DOI | Venue |
---|---|---|
2020 | 10.21437/Interspeech.2020-2213 | INTERSPEECH |
DocType | Citations | PageRank |
Conference | 1 | 0.36 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Lee Chi-Chang | 1 | 1 | 0.36 |
Lin Yu-Chen | 2 | 1 | 0.36 |
Hsuan-Tien Lin | 3 | 829 | 74.77 |
Hsin-min Wang | 4 | 1201 | 129.62 |
Yu Tsao | 5 | 208 | 50.09 |