Improving Short Utterance PLDA Speaker Verification using SUV Modelling and Utterance Partitioning Approach

TL;DR

This paper enhances short utterance speaker verification by partitioning long enrollment utterances into multiple segments and applying SUV modeling, leading to significant improvements in verification accuracy.

Contribution

It introduces a novel approach combining utterance partitioning with SUV modeling to improve PLDA-based speaker verification for short utterances.

Findings

Partitioning long utterances improves speaker verification accuracy.

SUV modeling compensates for mismatch between training and short test utterances.

Achieves 9% and 16% relative EER reduction on NIST datasets.

Abstract

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.