Please use this identifier to cite or link to this item: http://nopr.niscair.res.in/handle/123456789/11640
Title: Enhancement of noisy speech signal based on variance and modified gain function with PDE preprocessing technique for digital hearing aid
Authors: Deepa, D
Shanmugam, A
Keywords: MSE;PDE;SNR;Adaptive threshold;Gain function;Variance
Issue Date: May-2011
Publisher: NISCAIR-CSIR, India
Abstract: This paper deals with a single-channel speech enhancement algorithm, preprocessed using partial differential equation, used to overcome degradation of noisy speech signals in digital hearing aids. Adaptive threshold is estimated using variance in time index. Gain is modified based on adaptive threshold estimated in frequency bins and used to enhance noisy signal. By proposed method, definite improvement in SNR and reduced MSE can be obtained compared to conventional method. This method provides a greater degree of flexibility and control on noise subtraction levels that reduce artifacts in enhanced speech, resulting in improved speech quality and intelligibility.
Page(s): 332-337
URI: http://hdl.handle.net/123456789/11640
ISSN: 0975-1084 (Online); 0022-4456 (Print)
Appears in Collections:JSIR Vol.70(05) [May 2011]

Files in This Item:
File Description SizeFormat 
JSIR 70(5) 332-337.pdf107.69 kBAdobe PDFView/Open


Items in NOPR are protected by copyright, with all rights reserved, unless otherwise indicated.