Score calibration enables automatic speaker recognizers to make cost-effective accept / reject decisions. Traditional calibration requires supervised data, which is an expensive resource. We propose a 2-component GMM for unsupervised calibration and demonstrate good performance relative to a supervised baseline on NIST SRE'10 and SRE'12. A Bayesian analysis demonstrates that the uncertainty associated with the unsupervised calibration parameter estimates is surprisingly small.
@article{arxiv.1311.0707,
title = {Generative Modelling for Unsupervised Score Calibration},
author = {Niko Brümmer and Daniel Garcia-Romero},
journal= {arXiv preprint arXiv:1311.0707},
year = {2014}
}