Title :
Speech Denoising Based on Improved Discrete Wavelet Packet Decomposition
Author_Institution :
Chaohu Univ., Chaohu, China
Abstract :
The problem of speech denoising using wavelet packet decomposition algorithm is considered. For denoising, certain wavelet packet coefficients (frequency) are set to zero and the others are thresholded. We make a comparison of wavelet-packet coefficients algorithm and amended wavelet-packet frequency algorithm in the application of speech denoising. The amended wavelet-packet frequency algorithm is proposed and implemented that can further improve the performance of the denoising system.
Keywords :
discrete wavelet transforms; signal denoising; speech processing; amended wavelet-packet frequency algorithm; discrete wavelet packet decomposition algorithm; speech denoising; wavelet-packet coefficient algorithm; Discrete wavelet transforms; Noise reduction; Speech; Wavelet analysis; Wavelet packets; discrete wavelet packet decomposition; speech denoising; wavelet frequency denoising; wavelet packet coefficients;
Conference_Titel :
Network Computing and Information Security (NCIS), 2011 International Conference on
Conference_Location :
Guilin
Print_ISBN :
978-1-61284-347-6
DOI :
10.1109/NCIS.2011.182