Improved speech recognition using a subspace projection approach

Citation
Pc. Loizou et As. Spanias, Improved speech recognition using a subspace projection approach, IEEE SPEECH, 7(3), 1999, pp. 343-345
Citations number
10
Categorie Soggetti
Eletrical & Eletronics Engineeing
Journal title
IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING
ISSN journal
10636676 → ACNP
Volume
7
Issue
3
Year of publication
1999
Pages
343 - 345
Database
ISI
SICI code
1063-6676(199905)7:3<343:ISRUAS>2.0.ZU;2-7
Abstract
Two class-separability criteria based on the divergence measure are propose d to improve speech recognition performance. The average and weighted avera ge divergence measures are used as criteria for finding a transformation ma trix which maps the original features into a more discriminative subspace. Results are presented for a highly confusable task.