Saved in:
Bibliographic Details
Main Authors: Lu, Ye-Xin, Ai, Yang, Ling, Zhen-Hua
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2305.13686
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929207285645312
author Lu, Ye-Xin
Ai, Yang
Ling, Zhen-Hua
author_facet Lu, Ye-Xin
Ai, Yang
Ling, Zhen-Hua
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2305_13686
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle MP-SENet: A Speech Enhancement Model with Parallel Denoising of Magnitude and Phase Spectra
Lu, Ye-Xin
Ai, Yang
Ling, Zhen-Hua
Audio and Speech Processing
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.
title MP-SENet: A Speech Enhancement Model with Parallel Denoising of Magnitude and Phase Spectra
topic Audio and Speech Processing
url https://arxiv.org/abs/2305.13686