Speech Dereverberation Based on Variance-Normalized Delayed Linear Prediction

TLDR

The study introduces a statistical model‑based dereverberation method that cancels late reverberation from distant‑microphone recordings without requiring prior room impulse response knowledge. The method models the captured signal as a Gaussian source with time‑varying variance and a delayed linear prediction observation, optimizing a variance‑normalized sum of squared prediction errors to form the NDLP algorithm. NDLP robustly estimates an inverse late‑reverberation system in noisy conditions, preserves direct speech, improves performance with short observations, runs efficiently in the time‑frequency domain, and outperforms two existing approaches in experiments.

Abstract

This paper proposes a statistical model-based speech dereverberation approach that can cancel the late reverberation of a reverberant speech signal captured by distant microphones without prior knowledge of the room impulse responses. With this approach, the generative model of the captured signal is composed of a source process, which is assumed to be a Gaussian process with a time-varying variance, and an observation process modeled by a delayed linear prediction (DLP). The optimization objective for the dereverberation problem is derived to be the sum of the squared prediction errors normalized by the source variances; hence, this approach is referred to as variance-normalized delayed linear prediction (NDLP). Inheriting the characteristic of DLP, NDLP can robustly estimate an inverse system for late reverberation in the presence of noise without greatly distorting a direct speech signal. In addition, owing to the use of variance normalization, NDLP allows us to improve the dereverberation result especially with relatively short (of the order of a few seconds) observations. Furthermore, NDLP can be implemented in a computationally efficient manner in the time-frequency domain. Experimental results demonstrate the effectiveness and efficiency of the proposed approach in comparison with two existing approaches.