EUROSPEECH 2003  INTERSPEECH 2003

This paper introduces Locally Recurrent Probabilistic Neural Networks (LRPNN) as an extension of the wellknown Probabilistic Neural Networks (PNN). A LRPNN, in contrast to a PNN, is sensitive to the context in which events occur, and therefore, identification of time or spatial correlations is attainable. Besides the definition of the LRPNN architecture a fast threestep training method is proposed. The first two steps are identical to the training of traditional PNNs, while the third step is based on the Differential Evolution optimization method. Finally, the superiority of LRPNNs over PNNs on the task of textindependent speaker verification is demonstrated.
Bibliographic reference. Ganchev, Todor / Tasoulis, Dimitris K. / Vrahatis, Michael N. / Fakotakis, Nikos (2003): "Locally recurrent probabilistic neural network for textindependent speaker verification", In EUROSPEECH2003, 16731676.