Dual tree complex wavelet transform based speech enhancement

Abstract

Dual tree complex wavelet transform (DT-CWT) has the advantages of nearly shift-invariance and directional selectivity (for two or more dimensions) over the classical discrete wavelet transform. These advantages are essential for many signal processing applications (i.e. image fusion, image enhancement, pattern recognition). In his study, a speech enhancement method based on the DT-CWT has been proposed in order to test its performance in speech enhancement. An efficient estimator, multiplicatively modified log-spectral amplitude (MM-LSA) estimator is used for the enhancement of noisy subband wavelet coefficients. The objective and experimental results show the superiority of the proposed method to the wavelet transform based methods well known in the literature.