Short Utterance PLDA Speaker Verification using SN-WLDA and Variance Modelling Techniques

Ahilan, K.; Dean, D.; Sridharan, S.

dc.contributor.author	Ahilan, K.
dc.contributor.author	Dean, D.
dc.contributor.author	Sridharan, S.
dc.date.accessioned	2021-03-15T05:17:20Z
dc.date.accessioned	2022-06-27T10:02:19Z
dc.date.available	2021-03-15T05:17:20Z
dc.date.available	2022-06-27T10:02:19Z
dc.date.issued	2014
dc.identifier.citation	Kanagasundaram, A., Dean, D., & Sridharan, S. (2014). Short utterance PLDA speaker verification using SN-WLDA and variance modelling techniques. In Proceedings of the 15th Australasian International Conference on Speech Science and Technology (SST 2014) (pp. 155-158). New Zealand Institute of Language, Brain and Behaviour (NZILBB).	en_US
dc.identifier.uri	http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/1863
dc.description.abstract	This paper proposes a combination of source-normalized weighted linear discriminant analysis (SN-WLDA) and short utterance variance (SUV) PLDA modelling to improve the short utterance PLDA speaker verification. As short-length utterance i-vectors vary with the speaker, session variations and phonetic content of the utterance (utterance variation), a combined approach of SN-WLDA projection and SUV PLDA modelling is used to compensate the session and utterance variations. Experimental studies have found that a combination of SNWLDA and SUV PLDA modelling approach shows an improvement over baseline system (WCCN[LDA]-projected Gaussian PLDA (GPLDA)) as this approach effectively compensates the session and utterance variations.	en_US
dc.language.iso	en	en_US
dc.subject	speaker verification	en_US
dc.subject	session variation	en_US
dc.title	Short Utterance PLDA Speaker Verification using SN-WLDA and Variance Modelling Techniques	en_US
dc.type	Article	en_US