VACE-WPE: Virtual Acoustic Channel Expansion Based On Neural Networks for Weighted Prediction Error-Based Speech Dereverberation

TL;DR

This paper introduces VACE-WPE, a neural network-based framework that enhances single-microphone speech dereverberation by virtually expanding acoustic channels, significantly improving speech quality in noisy reverberant environments.

Contribution

The study extends the WPE algorithm with a neural network-based VACE framework, enabling effective dereverberation in single-microphone scenarios and simplifying the neural network training process.

Findings

VACE-WPE outperforms single-channel WPE in speech quality.

VACE-WPE is effective in noisy reverberant environments.

VACE-WPE enhances automatic speech recognition performance.

Abstract

Speech dereverberation is an important issue for many real-world speech processing applications. Among the techniques developed, the weighted prediction error (WPE) algorithm has been widely adopted and advanced over the last decade, which blindly cancels out the late reverberation component from the reverberant mixture of microphone signals. In this study, we extend the neural-network-based virtual acoustic channel expansion (VACE) framework for the WPE-based speech dereverberation, a variant of the WPE that we recently proposed to enable the use of dual-channel WPE algorithm in a single-microphone speech dereverberation scenario. Based on the previous study, some ablation studies are conducted regarding the constituents of the VACE-WPE in an offline processing scenario. These studies help understand the dynamics of the system, thereby simplifying the architecture and leading to the introduction of new strategies for training the neural network for the VACE. Experimental results in noisy reverberant environments reveal that VACE-WPE considerably outperforms its single-channel counterpart in terms of objective speech quality and is complementary to the single-channel WPE when employed as the front-end for the far-field automatic speech recognizer.