Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/76009
Title: Fast scoring for PLDA with uncertainty propagation via i-vector grouping
Authors: Lin, WW 
Mak, MW 
Chien, JT
Keywords: Speaker verification
I-Vector/PLDA
Uncertainty Propagation
Duration mismatch
Issue Date: 2017
Publisher: Academic Press
Source: Computer speech and language, 2017, v. 45, p. 503-515 How to cite?
Journal: Computer speech and language 
Abstract: The i-vector/PLDA framework has gained huge popularity in text-independent speaker verification. This approach, however, lacks the ability to represent the reliability of i-vectors. As a result, the framework performs poorly when presented with utterances of arbitrary duration. To address this problem, a method called uncertainty propagation (UP) was proposed to explicitly model the reliability of an i-vector by an utterance-dependent loading matrix. However, the utterance-dependent matrix greatly complicates the evaluation of likelihood scores. As a result, PLDA with UP, or PLDA-UP in short, is far more computational intensive than the conventional PLDA. In this paper, we propose to group i-vectors with similar reliability, and for each group the utterance-dependent loading matrices are replaced by a representative one. This arrangement allows us to pre-compute a set of representative matrices that cover all possible i-vectors, thereby greatly reducing the computational cost of PLDA-UP while preserving its ability in discriminating the reliability of i-vectors. Experiments on NIST 2012 SRE show that the proposed method can perform as good as the PLDA with UP while the scoring time is only 3.18% of it.
URI: http://hdl.handle.net/10397/76009
ISSN: 0885-2308
EISSN: 1095-8363
DOI: 10.1016/j.csl.2017.02.009
Appears in Collections:Journal/Magazine Article

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

SCOPUSTM   
Citations

3
Last Week
0
Last month
Citations as of Nov 12, 2018

WEB OF SCIENCETM
Citations

2
Last Week
0
Last month
Citations as of Nov 19, 2018

Page view(s)

22
Citations as of Nov 18, 2018

Google ScholarTM

Check

Altmetric


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