Fusion of face and speech data for person identity verification

Biometric person identity authentication is gaining more and more attention . The authentication task performed by an expert is a binary classification problem: reject or accept identity claim. Combining experts, each based on a different modality (speech, face, fingerprint, etc.), increases the perf ormance and robustness of identity authentication systems. In this context, a key issue is the fusion of the different experts for taking a. final dec ision (i.e., accept or reject identity claim). We propose to evaluate diffe rent binary classification schemes (support vector ;machine, multilayer per ceptron, C4.5 decision tree, Fisher's linear discriminant, Bayesian classif ier) to carry on the fusion, The experimental results show that support vec tor machines and Bayesian classifier achieve almost the same performances, and both outperform the other evaluated classifiers.