Deep Learning Based Phase Reconstruction for Speaker Separation: A Trigonometric Perspective
arXiv:1811.09010
Abstract
This study investigates phase reconstruction for deep learning based monaural talker-independent speaker separation in the short-time Fourier transform (STFT) domain. The key observation is that, for a mixture of two sources, with their magnitudes accurately estimated and under a geometric constraint, the absolute phase difference between each source and the mixture can be uniquely determined; in addition, the source phases at each time-frequency (T-F) unit can be narrowed down to only two candidates. To pick the right candidate, we propose three algorithms based on iterative phase reconstruction, group delay estimation, and phase-difference sign prediction. State-of-the-art results are obtained on the publicly available wsj0-2mix and 3mix corpus.
arXiv comment5 pages, in submission to ICASSP-2019
References in corpus (8)
- Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separation
- Speaker-independent Speech Separation with Deep Attractor Network
- Phasebook and Friends: Leveraging Discrete Representations for Source Separation
- ClariNet: Parallel Wave Generation in End-to-End Text-to-Speech
- TasNet: time-domain audio separation network for real-time, single-channel speech separation
- End-to-End Speech Separation with Unfolded Iterative Phase Reconstruction
- Improving DNN-based Music Source Separation using Phase Features
- Phase reconstruction from amplitude spectrograms based on von-Mises-distribution deep neural network