Tunable time delay neural networks for isolated word recognition

In this article, we describe a new neural network structure and a corresponding new sequential training technique for speech recognition. The proposed system is a modification of the original time delay neural network (TDNN) structure of Waibel (1989). The new structure consists of a group of sub-nets, and each isolated word to be recognized corresponds to at least one sub-net. Since each sub-net deals with only one word, it may be trained independently. Each sub-net is a TDNN which we train with a new sequential training algorithm. The system has attained close to 100% accuracy for a two-speaker, isolated word recognition task