The Research Of Speech Enhancement Method Based On Wavelet Transform And Real Reality

Presently, speech communication gains more and more widely application in variety of fields, meanwhile the reseach of speech enhancement technology plays very important role in improving the qulity of speech communication.A method of speech emhancement based on wavelet transform is researched in this paper. Above all, we introduce some speech enhancement algorithms and wavelet theory. Then we propose an speech enhancement methods based on adaptive wavelet threshold denoising method and called adaptive wavelet coefficients subtraction is proposed. The new speech enhancement method is much better than tranditional ones.The main idea of wavelet threshold lies in that when noise signal transforms from time domain to wavelet domain, the signal's wavelet coefficients will concentrate in a few area of wavelet domain, while the wavelet coefficients of noise will spread to all area of wavelet domain. Although the energy of noise is bigger than that of the signal, its wavelet coefficients are smaller than that of signal. So we can use threshold function to cut off the coefficients of noise and use the rest of coefficients to reconstruct the denoising signal.To satisfy the requirement of real-time processing for the time various signal, this paper proposes a new threshold function based on soft-threshold function. The new threshold function overcomes the shortcomings of traditional threshold function and has infinite order continuous derivative.Then based on SURE unbias estimation, LMS algorithm and the new threshold function, this paper proposes a speech enhancement method called adaptive wavelet coefficients subtraction. Using this method, the most optimal threshold parameters can be searched adaptively.It can satisfy the requirement of time various signal in real-time processing. The new speech enhancement method is much better than tranditional ones and has broad applications.At last, simulation of the algorithm based on Matlab software is implemented. The result of simulation proves that adaptive wavelet coefficients subtraction method can effectively restrain noise and increase SNR.The real-time implementation of the algorithm based on TMS320C54X DSP is verified.