DocumentCode :
2540996
Title :
Perceptually weighted multi-band spectral subtraction speech enhancement technique
Author :
Chowdhury, Md Faqrul Alam ; Alam, Md Jahangir ; Alam, Md Fasiul ; O´Shaughnessy, Douglas
Author_Institution :
Dept. of EEE, BUET, Dhaka
fYear :
2008
fDate :
20-22 Dec. 2008
Firstpage :
395
Lastpage :
399
Abstract :
It is gaining extensive research concentration to estimate a suitable factor that will subtract just the necessary amount of the noise spectrum from each frequency bin (ideally) to prevent destructive subtraction of the speech while removing most of the residual noise. Considering the advantage of multi-band spectral subtraction method which takes into account the fact that colored noise affects the speech spectrum differently at various frequencies and also of perceptual frequency weighting technique that allows for an automatic adaptation in the time and frequency of the enhancement system and finds a suitable noise estimate according to the frequency of the corrupted speech, our proposed method offers a better spectral approximation of the clean speech signal than these two method.
Keywords :
approximation theory; noise; speech enhancement; noise spectrum; perceptually weighted multiband spectral subtraction; residual noise removal; spectral approximation; speech enhancement technique; Additive noise; Colored noise; Electronic mail; Filters; Frequency estimation; Noise level; Signal processing; Speech coding; Speech enhancement; Working environment noise; MBSS; perceptual weighting filter; spectral subtraction; speech enhancement;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electrical and Computer Engineering, 2008. ICECE 2008. International Conference on
Conference_Location :
Dhaka
Print_ISBN :
978-1-4244-2014-8
Electronic_ISBN :
978-1-4244-2015-5
Type :
conf
DOI :
10.1109/ICECE.2008.4769239
Filename :
4769239
Link To Document :
بازگشت