Scene-Agnostic Multi-Microphone Speech Dereverberation

TL;DR

This paper introduces a neural network architecture capable of speech dereverberation that works with unknown and variable microphone array configurations, outperforming scene-aware and traditional methods in various conditions.

Contribution

The proposed neural network architecture handles variable and unknown microphone array configurations, advancing speech dereverberation techniques beyond fixed-array limitations.

Findings

Scene-agnostic model outperforms scene-aware frameworks with fewer microphones.

Method surpasses state-of-the-art WPE algorithm in noiseless conditions.

Effective on both noisy and noiseless reverberant datasets.

Abstract

Neural networks (NNs) have been widely applied in speech processing tasks, and, in particular, those employing microphone arrays. Nevertheless, most existing NN architectures can only deal with fixed and position-specific microphone arrays. In this paper, we present an NN architecture that can cope with microphone arrays whose number and positions of the microphones are unknown, and demonstrate its applicability in the speech dereverberation task. To this end, our approach harnesses recent advances in deep learning on set-structured data to design an architecture that enhances the reverberant log-spectrum. We use noisy and noiseless versions of a simulated reverberant dataset to test the proposed architecture. Our experiments on the noisy data show that the proposed scene-agnostic setup outperforms a powerful scene-aware framework, sometimes even with fewer microphones. With the noiseless dataset we show that, in most cases, our method outperforms the position-aware network as well as the state-of-the-art weighted linear prediction error (WPE) algorithm.