Paper Info
Title
Computationally Efficient Speaker Identification for Large Population Tasks using MLLR and Sufficient Statistics.
Abstract
In conventional Speaker-Identification using GMM-UBM framework, the likelihood of the given test utterance is computed with respect to all speaker-models before identifying the speaker, based on the maximum likelihood criterion. The calculation of likelihood score of the test utterance is computationally intensive, especially when there are tens of thousands of speakers in database. In this paper, we propose a computationally efficient (Fast) method to calculate the likelihood of the test utterance using speaker-specific Maximum Likelihood Linear Regression (MLLR) matrices (which are precomputed) and sufficient statistics estimated from the test utterance only once. We show that while this method is an order of magnitude faster, there is some degradation in performance. Therefore, we propose a cascaded system with the FastMLLR system identifying the top-N most probable speakers, followed by a conventional GMM-UBMto identify the most probable speaker from the top-N speakers. Experiments performed on the NIST 2004 database indicate that the cascaded system provides a speed up of 3.16 and 6.08 times for 1-side test (core condition) and 10 sec. test condition respectively, with a marginal degradation in accuracy over the conventional GMM-UBM system.
Year
Venue
Field
2010
ODYSSEY 2010: THE SPEAKER AND LANGUAGE RECOGNITION WORKSHOP
Population,Speaker identification,Pattern recognition,Computer science,Speech recognition,Artificial intelligence,Sufficient statistic
DocType
Citations 
PageRank 
Conference
0
0.34
References 
Authors
0
3
Name
Order
Citations
PageRank
Achintya Kumar Sarkar1237.81
Srinivasan Umesh29316.31
S. P. Rath3212.97