Generalized locally recurrent probabilistic neural networks for text-independent speaker verification

T. Ganchev, N. Fakotakis, D.K. Tasoulis, M.N. Vrahatis
2004 IEEE International Conference on Acoustics, Speech, and Signal Processing  
An extension of the well-known Probabilistic Neural Network (PNN), to Generalized Locally Recurrent PNN (GLR-PNN) is introduced. This extension renders GLRPNNs, in contrast to PNNs, sensitive to the context, in which events occur. A GLRPNN is therefore, able to identify time or spatial correlations. This capability can be exploited to improve performance on classification tasks. A fast three-step algorithm for training GLRPNNs is also proposed. The first two steps are identical to the training
more » ... f traditional PNNs, while the third step exploits the Differential Evolution optimization method. The performance of the proposed methodology on the task of text-independent speaker verification is contrasted with that of Locally Recurrent PNNs, Diagonal Recurrent Neural Networks, Infinite Impulse Response and Finite Impulse Response MLP-based structures, as well as with Gaussian Mixture Models-based classifier.
doi:10.1109/icassp.2004.1325917 dblp:conf/icassp/GanchevFTV04 fatcat:tcepk2x3dnh3hktawcvgiwl6q4