Research on the Denoising Algorithm of Speech Signal
Wavelet transform and Hilbert-Huang transform are the main methods on signal denoising. In this paper, the principle and classification of denoising methods based on wavelet transform are studied and the advantages and disadvantages of these methods are analyzed. At the same time, A denoising method of speech signal based on Hilbert-Huang transform and wavelet transform(HHT-WT) is proposed. Simulation experiments show that the HHT-WT method is a better speech denoising method.
Unable to display preview. Download preview PDF.
This work was supported by A Project of Shandong Province Higher Educational Science and Technology Program (J10LG20), China.
- 1.Man Jianfen, in: Application of Wavelet Transform in Speech Denoising, edited by volume 32 of Journal of Taiyuan university of technology (2001), p. 238–239.Google Scholar
- 2.Hou Bin, Du Guixian, Hu Min, in: Time-frequency spectral analysis of seismic data based on Hilbert-Huang transform, edited by volume 32 of Progress in exploration geophysics (2009), p. 248–251.Google Scholar
- 3.Huang N E,Zheng Shen,Long S R,et al, in: The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis, edited by volume 454 of Proceedings of the Royal Society of London (1998), p. 903–995.Google Scholar
- 4.Huang N E, Wu M C, Long S R,et al, in: A confidence limit for the empirical mode decomposition and Hilbert spectral analysis, edited by volume 459 of Proceedings of the Royal Society of London (2003), p. 2317–2345.Google Scholar
- 5.HUANGN E, SHEN Z, LONG S R, et al, in: The empirical mode decomposition and the Hilbert spectrum for nonlinear and nonstationary time series analysis [EB/OL. http://keck.ucsf.edu/~schenk/Huang_etal98 (2008).
- 6.I.M. Johnstone and B.W. Silverman, in: Wavelet threshold estimators for data with correlated noise. J. Roy. Statist. Soc. B. v01. (1997), p. 319–351.Google Scholar