Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/28185
Title: Boosting the performance of I-vector based speaker verification via utterance partitioning
Authors: Rao, W
Mak, MW 
Keywords: I-vectors
linear discriminant analysis
speaker verification
support vector machines
utterance partitioning with acoustic vector resampling (UP-AVR)
Issue Date: 2013
Publisher: Institute of Electrical and Electronics Engineers
Source: IEEE transactions on audio, speech and language processing, 2013, v. 21, no. 5, 6423258, p. 1012-1022 How to cite?
Journal: IEEE transactions on audio, speech and language processing 
Abstract: The success of the recent i-vector approach to speaker verification relies on the capability of i-vectors to capture speaker characteristics and the subsequent channel compensation methods to suppress channel variability. Typically, given an utterance, an i-vector is determined from the utterance regardless of its length. This paper investigates how the utterance length affects the discriminative power of i-vectors and demonstrates that the discriminative power of i-vectors reaches a plateau quickly when the utterance length increases. This observation suggests that it is possible to make the best use of a long conversation by partitioning it into a number of sub-utterances so that more i-vectors can be produced for each conversation. To increase the number of sub-utterances without scarifying the representation power of the corresponding i-vectors, repeated applications of frame-index randomization and utterance partitioning are performed. Results on NIST 2010 speaker recognition evaluation (SRE) suggest that (1) using more i-vectors per conversation can help to find more robust linear discriminant analysis (LDA) and within-class covariance normalization (WCCN) transformation matrices, especially when the number of conversations per training speaker is limited; and (2) increasing the number of i-vectors per target speaker helps the i-vector based support vector machines (SVM) to find better decision boundaries, thus making SVM scoring outperforms cosine distance scoring by 19% and 9% in terms of minimum normalized DCF and EER.
URI: http://hdl.handle.net/10397/28185
ISSN: 1558-7916
EISSN: 1558-7924
DOI: 10.1109/TASL.2013.2243436
Appears in Collections:Journal/Magazine Article

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

SCOPUSTM   
Citations

32
Last Week
0
Last month
0
Citations as of Aug 3, 2017

WEB OF SCIENCETM
Citations

24
Last Week
1
Last month
0
Citations as of Aug 21, 2017

Page view(s)

38
Last Week
1
Last month
Checked on Aug 21, 2017

Google ScholarTM

Check

Altmetric



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