Abstract not found

This paper considers the problems of estimating bigram language models and of efficiently representing them by a finite state network, which can be employed by an hidden Markov model based, beam-search, continuous speech recognizer.

An approximation of the Bayes decision rule and its implementation on a two-layered network are described. The net is trained in two phases: first, probabilities of the discrete-valued input features are learnt by applying a Good-Turing based estimator; second, net weights are estimated by applying an adaptive gradient descent technique. Experiments were performed on a database of 67,000 real life handwritten numerals. By using input units that read sub-patterns of the character bitmap, a recognition rate of 93.30% is achieved, with 1.39% substitution rate. The paper shows that computational complexity and implementation characteristics make this approach a possible competitor of artificial neural networks described in the literature. 1 Introduction Classification is the problem of mapping a set of patterns into a fixed number of classes. With a statistical approach, classification is usually based on an a posteriori probability. Moreover, many non-statistical classifiers can be seen...