5^{th} International Conference on Spoken Language ProcessingSydney, Australia |
For most classifier architectures realistic training schemes only allow classifiers corresponding to local optima of the training criteria to be constructed. One way of dealing with this problem is to work with classifier ensembles: multiple classifiers are trained for the same classification problem and combined into one ``super'' classifier. The problem addressed in this paper is text prompted speaker verification by means of phoneme dependent Radial Basis Function networks trained by gradient descent error minimisation. In this context ensemble techniques are introduced by combining different classifiers that classify feature vectors, which have been pre-processed using different linear transforms. Four different types of linear transforms are studied: the Fisher transform, the LDA transform, the PCA transform and the cosine transform. The verification system is evaluated on the Gandalf database, where the equal error rate is reduced from 3.6% to 3.2% when ensemble techniques are introduced.
Bibliographic reference. Olsen, Jesper Ostergaard (1998): "Speaker verification with ensemble classifiers based on linear speech transforms", In ICSLP-1998, paper 0334.