Short Utterance Variance Modelling and Utterance Partitioning for PLDA Speaker Verification

Ahilan Kanagasundaram, David Dean, Sridha Sridharan, Clinton Fookes, Ivan Himawan


This paper analyses the short utterance probabilistic linear discriminant analysis (PLDA) speaker verification with utterance partitioning and short utterance variance (SUV) modelling approaches. Experimental studies have found that instead of using single long-utterance as enrolment data, if long enrolled-utterance is partitioned into multiple short utterances and average of short utterance i-vectors is used as enrolled data, that improves the Gaussian PLDA (GPLDA) speaker verification. This is because short utterance i-vectors have speaker, session and utterance variations, and utterance-partitioning approach compensates the utterance variation. Subsequently, SUV-PLDA is also studied with utterance partitioning approach, and utterance-partitioning-based SUV-GPLDA system shows relative improvement of 9% and 16% in EER for NIST 2008 and NIST 2010 truncated 10sec-10sec evaluation condition as utterance-partitioning approach compensates the utterance variation and SUV modelling approach compensates the mismatch between full-length development data and short-length evaluation data.


DOI: 10.21437/Interspeech.2016-778

Cite as

Kanagasundaram, A., Dean, D., Sridharan, S., Fookes, C., Himawan, I. (2016) Short Utterance Variance Modelling and Utterance Partitioning for PLDA Speaker Verification. Proc. Interspeech 2016, 1835-1838.

Bibtex
@inproceedings{Kanagasundaram+2016,
author={Ahilan Kanagasundaram and David Dean and Sridha Sridharan and Clinton Fookes and Ivan Himawan},
title={Short Utterance Variance Modelling and Utterance Partitioning for PLDA Speaker Verification},
year=2016,
booktitle={Interspeech 2016},
doi={10.21437/Interspeech.2016-778},
url={http://dx.doi.org/10.21437/Interspeech.2016-778},
pages={1835--1838}
}