The application of Artificial Neural Networks (ANN) to Automatic Speech Recognition (ASR) is investigate in this thesis. The research in this thesis is oriented on the theory and application in the speech recognition of the Radial Basis Fuction Neural Networks(RBFNN), and the related algorithms and model are developed. The recognition of commands set is focused. The program design and the experiments are completed based on MATLAB7.0.1.In this thesis we discussed the methods of speech signal process on front port. The difficult methods of structure design and difficult learning algorithms of RBFNN to speech recognition are analyzed ,and the results are discussed. Through the further training method, the recognition capability of the RBFNN is improved much more. We adopt a sequential cluster method to solve the time alignment problem and discussed its effect on RBFNN. |