Enhancement of speech by using undecimated wavelet packet-perceptual filterbank and MM_LSA estimation

In this paper, we proposed a speech enhancement method which integrates a perceptual filterbank and multiplicatively modified log-spectral amplitude (MM_LSA) estimation. The perceptual filterbank was designed by adjusting undecimated wavelet packet decomposition tree, according to critical bands of human auditory system. The MM_LSA estimation was used for estimation of speech in undecimated wavelet packet domain. The perceptual filterbank provides a good auditory representation and perceptual quality of speech. Experimental results and objective quality measurement test results proved the performance of proposed speech enhancement method.