Back to Search
Start Over
A two-stage complex network using cycle-consistent generative adversarial networks for speech enhancement.
- Source :
-
Speech Communication . Nov2021, Vol. 134, p42-54. 13p. - Publication Year :
- 2021
-
Abstract
- Cycle-consistent generative adversarial networks (CycleGAN) have shown their promising performance for speech enhancement (SE), while one intractable shortcoming of these CycleGAN-based SE systems is that the noise components propagate throughout the cycle and cannot be completely eliminated. Additionally, conventional CycleGAN-based SE systems only estimate the spectral magnitude, while the phase is unaltered. Motivated by the multi-stage learning concept, we propose a novel two-stage denoising system that combines a CycleGAN-based magnitude enhancing network and a subsequent complex spectral refining network in this paper. Specifically, in the first stage, a CycleGAN-based model is responsible for only estimating magnitude, which is subsequently coupled with the original noisy phase to obtain a coarsely enhanced complex spectrum. After that, the second stage is applied to further suppress the residual noise components and estimate the clean phase by a complex spectral mapping network, which is a pure complex-valued network composed of complex 2D convolution/deconvolution and complex temporal-frequency attention blocks. Experimental results on two public datasets demonstrate that the proposed approach consistently surpasses previous one-stage CycleGANs and other state-of-the-art SE systems in terms of various evaluation metrics, especially in background noise suppression. • We propose a two-stage speech enhancement approach composed of a CycleGAN-based network and a complex denoising network. • We decompose the original complex spectrum estimation into two sub-tasks, i.e., maginitude and phase. • The proposed two-stage system outperforms the one-stage CycleGAN-based and many state-of-the-art speech enhancement approaches. [ABSTRACT FROM AUTHOR]
- Subjects :
- *GENERATIVE adversarial networks
*SPEECH enhancement
*CONCEPT learning
Subjects
Details
- Language :
- English
- ISSN :
- 01676393
- Volume :
- 134
- Database :
- Academic Search Index
- Journal :
- Speech Communication
- Publication Type :
- Academic Journal
- Accession number :
- 152794073
- Full Text :
- https://doi.org/10.1016/j.specom.2021.09.001