A Novel HMM and Wavelet Neural Network Hybrid Method for Noisy Speech Recognition

摘要

This paper presents a novel noisy speech recognition method based on hidden Markov model (HMM) and wavelet neural network (WNN) hybrid model. The HMM is employed to compute the Viterbi output vector. Then the vector is used as the inputs of WNN to acquire the new classify information. The result of recognition is made by fusion these two kinds of recognition information. Recognition experiment shows that this hybrid model has higher performance than hidden Markov model in noisy speech recognition.