DocumentCode :
2083057
Title :
Efficient speaker identification using distributional speaker model clustering
Author :
Apsingekar, Vijendra Raj ; De Leon, Phillip L.
Author_Institution :
Klipsch Sch. of Electr. & Comput. Eng., New Mexico State Univ., Las Cruces, NM
fYear :
2008
fDate :
26-29 Oct. 2008
Firstpage :
1260
Lastpage :
1264
Abstract :
For large population speaker identification (SI) systems, likelihood computations between an unknown speaker´s test feature vectors and speaker models can be very time-consuming and detrimental to applications where fast SI is required. In this paper, we propose a method whereby speaker models are clustered using a distributional distance measure such as KL divergence during the training stage. During the testing stage, only those clusters which are likely to contain high-likelihood speaker models are searched. The proposed method reduces the speaker model search space which directly results in faster SI. Any loss in identification accuracy can be controlled by trading off speed and accuracy. This paper implements GMM-UBM based SI system with MAP adapted speaker models and the results are presented on TIMIT, NTIMIT and NIST-2002 large population speech corpora.
Keywords :
biometrics (access control); speaker recognition; distributional distance; distributional speaker model clustering; high-likelihood speaker models; large population speaker identification systems; Application software; Biometrics; Distributed computing; Feature extraction; Loudspeakers; Security; Speaker recognition; Speech; System testing; Telephony;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signals, Systems and Computers, 2008 42nd Asilomar Conference on
Conference_Location :
Pacific Grove, CA
ISSN :
1058-6393
Print_ISBN :
978-1-4244-2940-0
Electronic_ISBN :
1058-6393
Type :
conf
DOI :
10.1109/ACSSC.2008.5074619
Filename :
5074619
Link To Document :
بازگشت