The Predictive Self-Organizing Map: application to speech features extraction

Abstract- Some well known theoretical results concerning the universal approximation property of MLP neural networks with one hidden layer have shown that for any function f from [0; 1]n to <, only the output layer weights depend on f. We use this result to propose a network architecture called the predictive Kohonen map allowing to design a new speech features extractor. We give experimental results of this approach on a phonemes recognition task. Key words- speech features extraction, function approximation, signal prediction