Enhancement of noisy speech based on decision-directed Wiener approach in perceptual wavelet packet domain
Md Tauhidul Islam, Mahtab Noor Shaan, Eashraque Jahan Easha, Ahmed Tahseen Minhaz, Celia Shahnaz, Shaikh Anowarul Fattah · 2017
In this paper, for enhancement of noisy speech, a method based on decision-directed Wiener approach in perceptual wavelet packet (PWP) domain is presented. The proposed method assumes an additive Gaussian noise model to derive the formulation of estimating the clean speech coefficients. The proposed method also considers the signal to noise ratio (SNR) information of the previous frame to obtain the estimates of the clean speech coefficients of the current frame. By using the speech files available in NOIZEUS database, a number of simulations are performed to evaluate the performance of the proposed method for speech signals in the presence of car, babble and street noises. The proposed method outperforms some of the state-of-the-art speech enhancement methods both at high and low levels of SNRs in terms of standard objective measures and subjective evaluations.