INTERSPEECH 2013
14thAnnual Conference of the International Speech Communication Association

Lyon, France
August 25-29, 2013

Likelihood-Ratio Calibration Using Prior-Weighted Proper Scoring Rules

Niko Brümmer (1), George R. Doddington (2)

(1) Agnitio, South Africa
(2) NULL, USA

Prior-weighted logistic regression has become a standard tool for calibration in speaker recognition. Logistic regression is the optimization of the expected value of the logarithmic scoring rule. We generalize this via a parametric family of proper scoring rules. Our theoretical analysis shows how different members of this family induce different relative weightings over a spectrum of applications of which the decision thresholds range from low to high. Special attention is given to the interaction between prior weighting and proper scoring rule parameters. Experiments on NIST SREf12 suggest that for applications with low false-alarm rate requirements, scoring rules tailored to emphasize higher score thresholds may give better accuracy than logistic regression.

Full Paper

Bibliographic reference.  Brümmer, Niko / Doddington, George R. (2013): "Likelihood-ratio calibration using prior-weighted proper scoring rules", In INTERSPEECH-2013, 1976-1980.