Confidence based multiple classifier fusion in speaker verification

A novel framework based on Bayes-based confidence measure for Multiple Classifier System fusion is proposed. Compared with ordinary Bayesian fusion, the presented approach leads to reductions of 20% and 25% in EER and ROC curve area, respectively, in speaker verification.

• Type: Journal
• Year: 2008
• Where: PRL
• Authors: Fernando Huenupán, Néstor Becerra Yoma, Carlos Molina, Claudio Garretón
• Keywords: Bayes-based Confidence Measure, Classifier System Fusion, Ordinary Bayesian Fusion, PRL 2008