Please use this identifier to cite or link to this item:
Title: Fast scoring for mixture of PLDA in i-vector/PLDA speaker verification
Authors: Mak, MW 
Issue Date: 2015
Source: 2015 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA), Hong Kong, China, 16-19 Dec 2015, p.587-593
Abstract: With the ubiquitous of mobile phones, users of speaker verification systems will perform authentication anywhere at anytime. As a result, practical speaker verification systems need to deal with utterances of different noise levels. Recently, an SNR-dependent mixture of PLDA model was proposed to deal with such practical situation. However, the scoring function of this model is significantly more complex than the conventional one. This paper proposes a method to reduce the computation burden of this mixture PLDA model. The idea is based on the observation that for most utterances, the posterior probabilities of SNR are very sparse so that it is possible to consider the top Gaussian only during scoring. The method effectively reduces the computational complexity from O(K2D3) to O(D3), where K and D are the number of mixtures and i-vector dimension, respectively. Experimental results based on NIST 2012 SRE suggest that the proposed method can reduce computation time by 60% with very minor degradation in performance.
Keywords: Signal to noise ratio
Computational complexity
Computational modeling
Noise measurement
Feature extraction
Analytical models
Publisher: Institute of Electrical and Electronics Engineers
ISBN: 978-9-8814-7680-7 (electronic)
978-1-4673-9593-9 (print on demand(PoD))
DOI: 10.1109/APSIPA.2015.7415337
Appears in Collections:Conference Paper

View full-text via PolyU eLinks SFX Query
Show full item record


Last Week
Last month
Citations as of Aug 14, 2020

Page view(s)

Last Week
Last month
Citations as of Sep 14, 2020

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.