Fig. 13From: A simulation study on optimal scores for speaker recognitionPerformance loss with the class mean scale mismatch between enrollment and test. The x-axis represents the scale factor αBack to article page