Wavelets Transformation Based Speech Signal Enhancement by Removing Impulsive Noise

Abstract

The presence of impulse noise in the speech signal has huge impact on the speech quality and on its performance in unprecedented levels. The removal of the impulse noise has been an area of research from over the years, but the work carried out in the past years fails to meet the desired requirements. A new approach based on the wavelets is proposed in this work to accomplish task of impulse noise removal in an accurate manner.  The wavelet based impulse noise removal system achieves better results than the traditional short time fourier transform (STFT) system, the utilization of the multi-resolution property of the wavelet transform provides good time resolution at the higher frequencies. It uses two features of speech to discriminate speech from impulse noise: one is the slow time-varying nature of speech and the other is the Lipschitz regularity of the speech components. On the basis of these features, an algorithm has been developed to identify and suppress wavelet coefficients that correspond to impulse noise. The simulation results provide an environment of preserving the speech quality while processing the noisy speech signal in order to remove the impulse noise from the speech.

Index Terms--- Speech signal, STFT, Wavelets, multi-resolution property, Speech quality, impulsive noise removal, speech enhancement

1. Introduction

The presence of impulse-like noise in speech can signifi- cantly reduces the intelligibility of speech and degrades automatic speech recognition performance. 

 Impulse noise is characterized by short bursts of acoustic energy having a wide spectral bandwidth and consisting of either isolated impulses or a series of impulses. Typical acoustic impulse noises include sounds of clicks in old phonograph recordings, of rain drops hitting a hard surface like the windshield