Dereverberation and denoising based on generalized spectral subtraction by multi-channel LMS algorithm using a small-scale microphone array

Longbiao Wang,Kyohei Odani,Atsuhiko Kai

doi:10.1186/1687-6180-2012-12

Longbiao Wang, Kyohei Odani + Show 1 more

Open Access

https://doi.org/10.1186/1687-6180-2012-12

Copy DOI

Abstract

A blind dereverberation method based on power spectral subtraction (SS) using a multi-channel least mean squares algorithm was previously proposed to suppress the reverberant speech without additive noise. The results of isolated word speech recognition experiments showed that this method achieved significant improvements over conventional cepstral mean normalization (CMN) in a reverberant environment. In this paper, we propose a blind dereverberation method based on generalized spectral subtraction (GSS), which has been shown to be effective for noise reduction, instead of power SS. Furthermore, we extend the missing feature theory (MFT), which was initially proposed to enhance the robustness of additive noise, to dereverberation. A one-stage dereverberation and denoising method based on GSS is presented to simultaneously suppress both the additive noise and nonstationary multiplicative noise (reverberation). The proposed dereverberation method based on GSS with MFT is evaluated on a large vocabulary continuous speech recognition task. When the additive noise was absent, the dereverberation method based on GSS with MFT using only 2 microphones achieves a relative word error reduction rate of 11.4 and 32.6% compared to the dereverberation method based on power SS and the conventional CMN, respectively. For the reverberant and noisy speech, the dereverberation and denoising method based on GSS achieves a relative word error reduction rate of 12.8% compared to the conventional CMN with GSS-based additive noise reduction method. We also analyze the effective factors of the compensation parameter estimation for the dereverberation method based on SS, such as the number of channels (the number of microphones), the length of reverberation to be suppressed, and the length of the utterance used for parameter estimation. The experimental results showed that the SS-based method is robust in a variety of reverberant environments for both isolated and continuous speech recognition and under various parameter estimation conditions.

Highlights

In a distant-talking environment, channel distortion drastically degrades speech recognition performance because of a mismatch between the training and testing environments
5.3 Experimental results of dereverberation and denoising reverberation and noise suppression using only 2 speech channels is described. c. In both spectral subtraction (SS)-based and generalized spectral subtraction (GSS)-based dereverberation methods, speech signals from two microphones were used to estimate blindly the compensation parameters for the power SS and GSS, and reverberation was suppressed by SS and the spectrum of dereverberant speech was inverted into a time domain
The one-stage dereverberation and denoising method based on GSS achieved a relative word error reduction rate of 12.8% compared to the conventional cepstral mean normalization (CMN) with GSS-based additive noise reduction method

Summary

Introduction

In a distant-talking environment, channel distortion drastically degrades speech recognition performance because of a mismatch between the training and testing environments. The current approach focusing on automatic speech recognition (ASR) robustness to reverberation and noise can be classified as speech signal processing, robust feature extraction, and model adaptation [1,2,3]. We focus on speech signal processing in the distant-talking environment. Because both the speech original (anechoic) speech. They applied a technique that they originally developed to treat background noise [7] to the dereverberation problem. A reverberation compensation method for speaker recognition using SS, in which late reverberation is treated as additive noise, was proposed in [9,10]. The drawback of this approach is that the optimum parameters for SS are empirically estimated from a development dataset and the late reverberation cannot be subtracted correctly as it is not modeled precisely

Methods

Results

Conclusion