Speech recognition based on wavelet packet transform and K-L expansion
Xu Wang, Zhiyan Han, Jian Wang, Yujuan Ma · 2008
Based on the dynamic characteristic of speech signal, we proposed a new method of number speech recognition using wavelet packet transform and K-L expansion. Firstly, speech signals underwent a series of preprocessing course including pre-filtering, quantification, pre-emphasizing and endpoint detector. Secondly, using wavelet packet transform extracted the relative energies in 32 sub-bands and the total energy of speech signals, then obtained as primary characteristic vector comprising 6 dimensions. Thirdly by K-L expansion, the primary characteristic vector with 33 dimensions was changed to that with 6 dimensions. Finally using BP networks as the classifier, the characteristic vector with 6 dimensions can maintain highly accurate recognition rate.