Simulation of Speech Denoising based on Voiced/ Unvoiced Decision by Using DWP
Basim Khalaf Jarullah1, Izz Kadhum Abboud2, Kareem Jabbar Tijil3
1Basim Khalaf Jarullah, University of Technology Electrical and Electronic Engineering Department, Iraq.
2Izz Kadhum Abboud, Mustansiriya University College of Engineering Computer And Software Engineering Department, Iraq.
3Kareem Jabbar Tijil, Institute of Technology Baghdad.
Manuscript received on November 27, 2013. | Revised Manuscript received on December 13, 2013. | Manuscript published on December 30, 2013. | PP: 230-233 | Volume-3, Issue-2, December 2013. | Retrieval Number: B2447123213/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In this paper, a speech denoising system using wavelet packet thresholding algorithm is investigated. This method is based on thresholding the wavelet coefficients that depend on voiced/unvoiced detection. Wavelet threshold can be done by a standard deviation method for each frame by level dependent thresholding using semisoft threshold in the additive Gaussian noise channel. The results of simulation indicate that using Discrete Wavelet Packet (DWP) in speech denoising application provides a quality better than Discrete Wavelet Thresholding (DWT) and voiced/unvoiced detection enhanced the performance of the system.
Keywords: DWP, DWT, Matlab.