In this paper we present a speaker recognition algorithm that models explicitly intra-speaker inter-session variability. Such variability may be caused by changing speaker characteristics (mood, fatigue, etc.), channel variability or noise variability. We define a session-space in which each session (either train or test session) is a vector. We then calculate a rotation of the session-space for which the estimated intra-speaker subspace is isolated and can be modeled explicitly. We evaluated our technique on the NIST-2004 speaker recognition evaluation corpus, and compared it to a GMM baseline system. Results indicate significant reduction in error rate.
Cite as: Aronowitz, H., Irony, D., Burshtein, D. (2005) Modeling intra-speaker variability for speaker recognition. Proc. Interspeech 2005, 2177-2180, doi: 10.21437/Interspeech.2005-643
@inproceedings{aronowitz05_interspeech,
author={Hagai Aronowitz and Dror Irony and David Burshtein},
title={{Modeling intra-speaker variability for speaker recognition}},
year=2005,
booktitle={Proc. Interspeech 2005},
pages={2177--2180},
doi={10.21437/Interspeech.2005-643}
}