Speaker recognition using continuous density support vector machines

Authors
Xin, D Wu, ZH
Citation
D. Xin et Zh. Wu, Speaker recognition using continuous density support vector machines, ELECTR LETT, 37(17), 2001, pp. 1099-1101
Citations number
4
Categorie Soggetti
Eletrical & Eletronics Engineeing
Journal title
ELECTRONICS LETTERS
ISSN journal
00135194 → ACNP
Volume
37
Issue
17
Year of publication
2001
Pages
1099 - 1101
Database
ISI
SICI code
0013-5194(20010816)37:17<1099:SRUCDS>2.0.ZU;2-G
Abstract
A new classification system for text-independent speaker recognition is pre sented. This system combines the output probabilities of generative Gaussia n mixture models and discriminative classifier support vector machines. The new method is tested on the YOHO database and achieves superior performanc e.