Fusion of the base classifiers is seen as the way to achieve state-ofthe art performance in the speaker verification systems. Standard approach is to pose the fusion problem as the linear binary classification task. Most successful loss function in speaker verification fusion has been the weighted logistic regression popularized by the FoCal toolkit. However, it is known that optimizing logistic regression can overfit severely without appropriate regularization. In addition, subset classifier selection can be achieved by using an external 0/1 loss function on the best subset. In this work, we propose to use LASSO based regularization on the FoCal cost function to achieve improved performance and classifier subset selection method integrated into one optimization task. Proposed method is able to achieve 51% relative improvement in Actual DCF over the FoCal baseline.
Bibliographic reference. Hautamäki, Ville / Lee, Kong Aik / Kinnunen, Tomi / Ma, Bin / Li, Haizhou (2011): "Regularized logistic regression fusion for speaker verification", In INTERSPEECH-2011, 2745-2748.