NESSR : a Neural Expert System for Speech Recognition

and key words Artificial neural networks (ANNs) have found applications in large spectrum of fields. Satisfactory results are obtained particularly in classification problems. In speech recognition context, the use of ANNs is hard, this is essentially due to the absence of the temporal aspect in their structure. On the other hand, assuming a speech recognition task, a word could be recognized and well categorized or recognized and badly categorized ; so the explanation of the decision is very important. In this paper, we address two limitations of ANNs : the lack of explicit knowledge and the absence of temporal aspect in their implementation. STN : is a model of a specialized temporal neuron, which includes both symbolic and temporal aspects. To illustrate the STN utility, we consider a system for speech recognition ; we underline in this paper the explanation aspect of the system. Artificial intelligence, machine learning, connectionism, speech recognition. traitement du signal 2007_volume 24_numéro 1 59