Toward optimizing stream fusion in multistream recognition of speech
Open Access
- 16 June 2011
- journal article
- research article
- Published by Acoustical Society of America (ASA) in The Journal of the Acoustical Society of America
- Vol. 130 (1) , EL14-EL18
Abstract
A multistream phoneme recognition framework is proposed based on forming streams from different spectrotemporal modulations of speech. Phoneme posterior probabilities were estimated from each stream separately and combined at the output level. A statistical model of the final estimated posterior probabilities is used to characterize the system performance. During the operation, the best fusion architecture is chosen automatically to maximize the similarity of output statistics to clean condition. Results on phoneme recognition from noisy speech indicate the effectiveness of the proposed method.Keywords
This publication has 8 references indexed in Scilit:
- Phoneme representation and classification in primary auditory cortexThe Journal of the Acoustical Society of America, 2008
- Denoising in the Domain of Spectrotemporal ModulationsEURASIP Journal on Audio, Speech, and Music Processing, 2007
- Combination of Acoustic Classifiers Based on Dempster-Shafer Theory of Evidence2007
- Multiresolution spectrotemporal analysis of complex soundsThe Journal of the Acoustical Society of America, 2005
- Multi-band speech recognition in noisy environments2002
- Connectionist Speech Recognition1994
- Perceptual linear predictive (PLP) analysis of speechThe Journal of the Acoustical Society of America, 1990
- Thirteen Ways to Look at the Correlation CoefficientThe American Statistician, 1988