Implementation and analysis of training algorithms for the classification of infant cry with feed-forward neural networks

We present the development of an automatic recognition system of infant cry, with the objective to classify two types of cry: normal and pathological cry from deaf babies. We used acoustic characteristics obtained by the linear prediction technique and as a classifier a feedforward neural network that was trained with several learning methods, resulting better the scaled conjugate gradient algorithm. Current results are shown, which, up to the moment, are very encouraging with an accuracy up to 94.3%.