Speech Signal Enhancement Techniques

Zegar, Chouki; Dahimene, Abdelhakim

AL-Lisaniyyat
Volume 20, Numéro 1, Pages 63-76
2014-06-28

Speech Signal Enhancement Techniques

Authors : Zegar Chouki . Dahimene Abdelhakim .

Abstract

Speech enhancement and noise reduction have wide applications in speech processing. They are often employed as preprocessing stage in various applications. The work to be presented in this paper is denoising a single-channel speech signal in the presence of a highly non-stationary background noise in order to improve the perceptible quality and intelligibility of the speech. Real world noise is mostly highly non-stationary and does not affect the speech signal uniformly over the spectrum. This paper explores a set of DFT-based algorithms as single-channel speech enhancement techniques which are as follows:  Spectral Subtraction using oversubtraction and spectral floor.  Multi-Band Spectral Subtraction (MBSS).  Wiener Filter.  MMSE of Short-Time Spectral Amplitude (MMSE-STSA) estimator with, and without using SPU modifier.  MMSE Log-Spectral Amplitude Estimator with, and without using SPU modifier. Optimally-Modified Log-Spectral Amplitude estimator (OM-LSA). The comparison study results based on subjective and objective tests showed that the Optimally Modified Log-Spectral Amplitude Estimator (OM-LSA) method outperforms all the implemented DFTbased single-channel speech enhancement algorithms