Low Latency Online Blind Source Separation Based on Joint Optimization with Blind Dereverberation

Abstract

This paper presents a new low-latency online blind source separation (BSS) algorithm. Although algorithmic delay of a frequency domain online BSS can be reduced simply by shortening the shorttime Fourier transform (STFT) frame length, it degrades the source separation performance in the presence of reverberation. This paper proposes a method to solve this problem by integrating BSS with Weighted Prediction Error (WPE) based dereverberation. Although a simple cascade of online BSS after online WPE upgrades the separation performance, the overall optimality is not guaranteed. Instead, this paper extends a recently proposed batch processing algorithm that can jointly optimize dereverberation and separation so that it can perform online processing with low computational cost and little processing delay (< 12 ms). The results of a source separation experiment in a noisy car environment suggest that the proposed online method has better separation performance than the simple cascaded methods.

Causal and Relaxed-Distortionless Response Beamforming for Online Target Source Extraction

“…The long window allows us to assume the instantaneous mixing process, and the short window reduces the algorithmic delay. Another approach considers the convolutive mixing process in the T-F domain and uses short windows in both STFT and iSTFT [16], [17]. Although various T-F domain approaches have been proposed [16], [17], [20]- [22], they cannot reduce the algorithmic delay to zero owing to the use of STFT.…”