Title :
The Improved VQ Algorithm for Speaker Recognition
Author :
Huo Chunbao ; Shao Yan ; Gao XuGuang
Author_Institution :
Electron. & Inf. Eng. Coll., Liaoning Univ. of Technol., Jinzhou, China
Abstract :
VQ is a universal algorithm for speaker recognition. In this paper, we propose an improved VQ algorithm with the new weighted measure, which takes into account the correlations between inter- and intra- space of vectors. A series of Matlab experiments are implemented, and the experiment results show that the new VQ algorithm enhances the speaker difference and boosts the rate of the speaker recognition.
Keywords :
speaker recognition; vector quantisation; Matlab; improved VQ algorithm; speaker difference; speaker recognition; Books; Educational institutions; Electronic equipment testing; Extraterrestrial measurements; Lagrangian functions; Partitioning algorithms; Size measurement; Space technology; Speaker recognition; Weight measurement;
Conference_Titel :
Innovative Computing, Information and Control (ICICIC), 2009 Fourth International Conference on
Conference_Location :
Kaohsiung
Print_ISBN :
978-1-4244-5543-0
DOI :
10.1109/ICICIC.2009.357
