Hybrid HMM/DTW based Speech Recognition with Kernel Adaptive Filtering Method

Abstract

We have proposed new approach for the speech recognition system by applying kernel adaptive filterfor speech enhancement and for the recognition,the hybrid HMM/DTW methods are usedin this paper. Noise removal is very important in many applications like telephone con versation, speech recognition, etc.In the recent past, the kernel methods are showing good results for speech processing applications . The feature used in the recognition process is MFCC features. It consists of a HMM system used to trainthe speech features and for classification purpose used the DTW method. Experimental results show a relative improvementof recognition rate compared to the traditional methods.