The Research On Speech Enhancement Based On Wavelet Packet Transform Algorithms

In many speech processing applications, it is very common to find the degradation of the quality of speech caused by undesirable background noise. The goal of speech enhancement is to recover original speech signals from noisy observations.Wavelet theory is a newly developed time-frequency analysis technique and is especially of interest for the analysis of non-stationary signal such as speech, sonar seismic signal, etc. The main idea of wavelet thresholding lies in that when noising signal transforms from time domain to wavelet domain, the signal's wavelet coefficients will spread to all area of wavelet domain. Although the energy of noise is bigger than the signal, its wavelet coefficients are smaller than the signal's. So we can use thresholding function to cut off the coefficients of noise and use the rest of coefficients to reconstruct the denoising signal.This paper presents a new algorithm for speech enhancement based on wavelet thresholding method. First, we decompose the noisy speech by the Bark-scaled Wavelet Packet (BS-WPD) to simulate the human auditory characteristics. Then we propose a new thresholding function which has many advantages over soft and hard thresholdings put forward by D.L. Donoho and I.M. Johnstone. Simulation result proves that the proposed new thresholding function has a better improvement.At last, simulation of the algorithm based on Matlab software is implemented. A large amount of simulation results indicate that our new method based on Bark-scaled wavelet packet decomposition has a better performance in both objective and subjective aspects.