ISCA Archive Interspeech 2005
ISCA Archive Interspeech 2005

Modelling session variability in text-independent speaker verification

Robbie Vogt, Brendan Baker, Sridha Sridharan

Presented is an approach to modelling session variability for GMM-based text-independent speaker verification incorporating a constrained session variability component in both the training and testing procedures. The proposed technique reduces the data labelling requirements and removes discrete categorisation needed by techniques such as feature mapping and H-Norm, while providing superior performance. Experiments on Switchboard-II conversational telephony data show improvements of as much as 48% in detection cost with a single training utterance and 68% with multiple training utterances over a baseline system.

doi: 10.21437/Interspeech.2005-670

Cite as: Vogt, R., Baker, B., Sridharan, S. (2005) Modelling session variability in text-independent speaker verification. Proc. Interspeech 2005, 3117-3120, doi: 10.21437/Interspeech.2005-670

  author={Robbie Vogt and Brendan Baker and Sridha Sridharan},
  title={{Modelling session variability in text-independent speaker verification}},
  booktitle={Proc. Interspeech 2005},