In this paper, a new speech enhancement method is introduced. It is essentially based on the Empirical Mode Decomposition technique (EMD) and a soft thresholding approach applied on selected modes. The proposed method is a fully data driven approach. First the noisy speech signal is decomposed adaptively into intrinsic oscillatory components called Intrinsic Mode Functions (IMFs) by using a time decomposition called sifting process. Second, selected IMFs are soft thresholded and added to the remaining IMFs with the residue to reconstitute the enhanced speech signal. The proposed approach is evaluated using speech signals from NOISEUS database corrupted with additive white Gaussian noise. Our algorithm is compared to other state of the art algorithms.
Copyrights (c) 2012 Computer Science Journals. All rights reserved.
Best viewed at 1152 x 864 resolution. Microsoft Internet Explorer.
Copyrights & Usage: Articles published by CSC Journals are Open Access. Permission to copy and distribute any other content, images, animation and other parts of this website is prohibited. CSC Journals has the rights to take action against individual/group if they are found victim of copying these parts of the website.