Phase-aware Speech Enhancement with Deep Complex U-Net

by   Hyeong-Seok Choi, et al.

Most deep learning-based models for speech enhancement have mainly focused on estimating the magnitude of spectrogram while reusing the phase from noisy speech for reconstruction. This is due to the difficulty of estimating the phase of clean speech. To improve speech enhancement performance, we tackle the phase estimation problem in three ways. First, we propose Deep Complex U-Net, an advanced U-Net structured model incorporating well-defined complex-valued building blocks to deal with complex-valued spectrograms. Second, we propose a polar coordinate-wise complex-valued masking method to reflect the distribution of complex ideal ratio masks. Third, we define a novel loss function, weighted source-to-distortion ratio (wSDR) loss, which is designed to directly correlate with a quantitative evaluation measure. Our model was evaluated on a mixture of the Voice Bank corpus and DEMAND database, which has been widely used by many deep learning models for speech enhancement. Ablation experiments were conducted on the mixed dataset showing that all three proposed approaches are empirically valid. Experimental results show that the proposed method achieves state-of-the-art performance in all metrics, outperforming previous approaches by a large margin.


page 14

page 18


Phase Aware Speech Enhancement using Realisation of Complex-valued LSTM

Most of the deep learning based speech enhancement (SE) methods rely on ...

Phase-aware Single-stage Speech Denoising and Dereverberation with U-Net

In this work, we tackle a denoising and dereverberation problem with a s...

Real-time Monaural Speech Enhancement With Short-time Discrete Cosine Transform

Speech enhancement algorithms based on deep learning have been improved ...

Differentiable Consistency Constraints for Improved Deep Speech Enhancement

In recent years, deep networks have led to dramatic improvements in spee...

PHASEN: A Phase-and-Harmonics-Aware Speech Enhancement Network

Time-frequency (T-F) domain masking is a mainstream approach for single-...

Complex spectrogram enhancement by convolutional neural network with multi-metrics learning

This paper aims to address two issues existing in the current speech enh...

Real-time Denoising and Dereverberation with Tiny Recurrent U-Net

Modern deep learning-based models have seen outstanding performance impr...