Speech Denoising Based on Wavelet Transform and Wiener Filtering

This paper presents a combinational method of Wavelet Transform and Adaptive Filter to reduce distortion that has been caused by interference noise. The procedure makes use of a Wiener filtering to improve and enhance the quality of the speech signal. The proposed method is tested for a couple of noisy speech signals with different SNRs in the range of −5 dB to 15 dB. In this method we aim to inspect the enhancement of noisy speech signals in more practical conditions. Thus, not only is the process employed on noisy speech signals distorted by Additive White Gaussian Noise (WGN) but also it is practiced on environmental noises such as CAR noise, etc. The results show improvement of noise reduction after applying Wiener filtering to the Wavelet Transform compared with utilizing Wavelet Transform only.