multiband spectral subtraction 2009-02-27 09:19:35 About multiband spectral subtraction, I have a question. According to the literature of multiband spectral subtraction ,written by Sunil D. Kamath and Philipos C. Loizou , I conclude the idea of that article. Actually, it use the method of Berouti. It spilt frequency band into several subbands, apply the method of berouti in per subband. And they also think it should have 4 subbands. More subbands are not necessary due to trivial increase of speech quality. At the same time,they think the method of berouti is not band-splited ,i.e ,only one subband. But I read the literature of Berouti. The method is actually band-splited. But Berouti did not say it literally and directly. In a case of 256 points fft,it split 129 sub-bands. Berouti did everything other than use the term of multiband. 1. Enhancement of speech corrupted by acoustic noiseBerouti, M. Schwartz, R. Makhoul, J. Bolt Beranek and Newman, Inc., Cambridge, MA;2. S. Kamath and P. Loizou. “A multi-band spectral subtraction method for enhancing speech corrupted by colored noise”, In Proceedings International Conference on Acoustics, Speech and Signal Processing, 2002 Any comments would be appreciated! Cheers HyeeWang@gmail.com Last Edit: 2009-02-27 14:55:17 by hyeewang