The technique known as bootstrapping or resampling has been used effectively in the field of statistics to obtain good estimates of statistics from only a small set of observations. In this paper we explore the use of this powerful technique to aid in improving the performance of a GMM-UBM text-independent speaker recognition system. We apply the bootstrap to the training process in the generation of speaker models for the GMM-UBM system. We also aggregate the outputs of the bootstraps multiple speaker models in our bagging system. Speaker recognition results of our bootstrap and bagging systems are presented on NIST corpora.
Cite as: Andrews, W.D., Campbell, J.P., Reynolds, D.A. (2000) Bootstrapping for speaker recognition. Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000), vol. 2, 483-486, doi: 10.21437/ICSLP.2000-312
@inproceedings{andrews00_icslp,
author={Walter D. Andrews and Joseph P. Campbell and Douglas A. Reynolds},
title={{Bootstrapping for speaker recognition}},
year=2000,
booktitle={Proc. 6th International Conference on Spoken Language Processing (ICSLP 2000)},
pages={vol. 2, 483-486},
doi={10.21437/ICSLP.2000-312}
}