MP-SENet: A Speech Enhancement Model with Parallel Denoising of Magnitude and Phase Spectra
arXiv:2305.13686 · doi:10.21437/Interspeech.2023-1441
Abstract
This paper proposes MP-SENet, a novel Speech Enhancement Network which directly denoises Magnitude and Phase spectra in parallel. The proposed MP-SENet adopts a codec architecture in which the encoder and decoder are bridged by convolution-augmented transformers. The encoder aims to encode time-frequency representations from the input noisy magnitude and phase spectra. The decoder is composed of parallel magnitude mask decoder and phase decoder, directly recovering clean magnitude spectra and clean-wrapped phase spectra by incorporating learnable sigmoid activation and parallel phase estimation architecture, respectively. Multi-level losses defined on magnitude spectra, phase spectra, short-time complex spectra, and time-domain waveforms are used to train the MP-SENet model jointly. Experimental results show that our proposed MP-SENet achieves a PESQ of 3.50 on the public VoiceBank+DEMAND dataset and outperforms existing advanced speech enhancement methods.
Accepted by Interspeech 2023
Cited by in corpus (5)
- Bilingual Dual-Head Deep Model for Parkinson's Disease Detection from Speech
- TF-CorrNet: Leveraging Spatial Correlation for Continuous Speech Separation
- CleanUMamba: A Compact Mamba Network for Speech Denoising using Channel Pruning
- Study of Lightweight Transformer Architectures for Single-Channel Speech Enhancement
- A Dual-Branch Parallel Network for Speech Enhancement and Restoration