Title :
Improved speech recognition using a subspace projection approach
Author :
Loizou, Philipos C. ; Spanias, Andreas S.
Author_Institution :
Dept. of Appl. Sci., Arkansas Univ., Little Rock, AR, USA
fDate :
5/1/1999 12:00:00 AM
Abstract :
Two class separability criteria based on the divergence measure are proposed to improve speech recognition performance. The average and weighted average divergence measures are used as criteria for finding a transformation matrix which maps the original features into a more discriminative subspace. Results are presented for a highly confusable task
Keywords :
matrix algebra; probability; speech recognition; average divergence measures; class separability criteria; divergence measure; highly confusable task; speech recognition; subspace projection approach; transformation matrix; weighted average divergence measures; Covariance matrix; Degradation; Density functional theory; Hidden Markov models; Maximum likelihood estimation; Optimization methods; Speech recognition; Telephony; Vocabulary; Weight measurement;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on
