Complex Spectral Mapping with a Convolutional Recurrent Network for Monaural Speech Enhancement
Authors:
Top Cited Papers
- 1 May 2019
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 6865-6869
Abstract
Phase is important for perceptual quality in speech enhancement. However, it seems intractable to directly estimate phase spectrogram through supervised learning due to lack of clear structure in phase spectrogram. Complex spectral mapping aims to estimate the real and imaginary spectrograms of clean speech from those of noisy speech, which simultaneously enhances magnitude and phase responses of noisy speech. In this paper, we propose a new convolutional recurrent network (CRN) for complex spectral mapping, which leads to a causal system for noise- and speaker-independent speech enhancement. In terms of objective intelligibility and perceptual quality, the proposed CRN significantly outperforms an existing convolutional neural network (CNN) for complex spectral mapping, as well as a strong CRN for magnitude spectral mapping. We additionally incorporate a newly-developed group strategy to substantially reduce the number of trainable parameters and the computational cost without sacrificing performance.Keywords
This publication has 16 references indexed in Scilit:
- Long short-term memory for speaker generalization in supervised speech separationThe Journal of the Acoustical Society of America, 2017
- Complex Ratio Masking for Monaural Speech SeparationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2015
- STFT Phase Reconstruction in Voiced Speech for an Improved Single-Channel Speech EnhancementIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- On Training Targets for Supervised Speech SeparationIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2014
- Learning spectral mapping for speech dereverberation2014
- Deep learning for monaural speech separation2014
- The importance of phase in speech enhancementSpeech Communication, 2011
- Iterative Phase Estimation for the Synthesis of Separated Sources From Single-Channel MixturesIEEE Signal Processing Letters, 2010
- The design for the wall street journal-based CSR corpus1992
- The unimportance of phase in speech enhancementIEEE Transactions on Acoustics, Speech, and Signal Processing, 1982