DocumentCode :
1436671
Title :
Analysis of Large-Scale SVM Training Algorithms for Language and Speaker Recognition
Author :
Cumani, Sandro ; Laface, Pietro
Author_Institution :
Dipt. di Autom. e Inf., Politec. di Torino, Torino, Italy
Volume :
20
Issue :
5
fYear :
2012
fDate :
7/1/2012 12:00:00 AM
Firstpage :
1585
Lastpage :
1596
Abstract :
This paper compares a set of large scale support vector machine (SVM) training algorithms for language and speaker recognition tasks. We analyze five approaches for training phonetic and acoustic SVM models for language recognition. We compare the performance of these approaches as a function of the training time required by each of them to reach convergence, and we discuss their scalability towards large corpora. Two of these algorithms can be used in speaker recognition to train a SVM that classifies pairs of utterances as either belonging to the same speaker or to two different speakers. Our results show that the accuracy of these algorithms is asymptotically equivalent, but they have different behavior with respect to the time required to converge. Some of these algorithms not only scale linearly with the training set size, but are also able to give their best results after just a few iterations. State-of-the-art performance has been obtained in the female subset of the NIST 2010 Speaker Recognition Evaluation extended core test using a single SVM system.
Keywords :
natural language processing; speaker recognition; support vector machines; NIST 2010 Speaker Recognition Evaluation; acoustic SVM model; language recognition; large-scale SVM training algorithms; phonetic SVM model; speaker recognition; support vector machine; Acoustics; Kernel; Speaker recognition; Speech processing; Support vector machines; Training; Vectors; Language recognition; large-scale training; speaker recognition; support vector machines (SVMs);
fLanguage :
English
Journal_Title :
Audio, Speech, and Language Processing, IEEE Transactions on
Publisher :
ieee
ISSN :
1558-7916
Type :
jour
DOI :
10.1109/TASL.2012.2186290
Filename :
6143993
Link To Document :
بازگشت