Low algorithmic delay implementation of convolutional beamformer for online joint source separation and dereverberation

TL;DR

This paper introduces a low-latency convolutional beamformer for real-time joint source separation and dereverberation, integrating WPE with NST-IVA to reduce delay while maintaining high separation quality.

Contribution

It presents a novel integration of WPE with NST-IVA, achieving low algorithmic delay and improved separation performance for online audio source separation.

Findings

Significantly reduced latency compared to existing online algorithms.

Validated improved separation performance through simulations.

Maintains low delay by controlling WPE processing.

Abstract

Blind-audio-source-separation (BASS) techniques, particularly those with low latency, play an important role in a wide range of real-time systems, e.g., hearing aids, in-car hand-free voice communication, real-time human-machine interaction, etc. Most existing BASS algorithms are deduced to run on batch mode, and therefore large latency is unavoidable. Recently, some online algorithms were developed, which achieve separation on a frame-by-frame basis in the short-time-Fourier-transform (STFT) domain and the latency is significantly reduced as compared to those batch methods. However, the latency with these algorithms may still be too long for many real-time systems to bear. To further reduce latency while achieving good separation performance, we propose in this work to integrate a weighted prediction error (WPE) module into a non-causal sample-truncating-based independent vector analysis (NST-IVA). The resulting algorithm can maintain the algorithmic delay as NST-IVA if the delay with WPE is appropriately controlled while achieving significantly better performance, which is validated by simulations.