TaylorBeamixer: Learning Taylor-Inspired All-Neural Multi-Channel Speech Enhancement from Beam-Space Dictionary Perspective

TL;DR

TaylorBM is an all-neural beamformer inspired by Taylor series expansion, which models beam-space components for improved speech enhancement, demonstrating superior performance on LibriSpeech data.

Contribution

It introduces a novel neural beamformer that simulates Taylor series expansion to enhance multi-channel speech signals end-to-end.

Findings

Outperforms existing advanced baselines in evaluation metrics

Effective modeling of beam-space components improves speech enhancement

End-to-end system demonstrates promising results on LibriSpeech corpus

Abstract

Despite the promising performance of existing frame-wise all-neural beamformers in the speech enhancement field, it remains unclear what the underlying mechanism exists. In this paper, we revisit the beamforming behavior from the beam-space dictionary perspective and formulate it into the learning and mixing of different beam-space components. Based on that, we propose an all-neural beamformer called TaylorBM to simulate Taylor's series expansion operation in which the 0th-order term serves as a spatial filter to conduct the beam mixing, and several high-order terms are tasked with residual noise cancellation for post-processing. The whole system is devised to work in an end-to-end manner. Experiments are conducted on the spatialized LibriSpeech corpus and results show that the proposed approach outperforms existing advanced baselines in terms of evaluation metrics.