Speech recognition system using adapted local trigonometric transforms
Sungwook Chang, Y. Kwon, Sung-II Yang · 2002
We propose a new speech feature extracted from adapted local trigonometric transforms. We decompose speech signal through adapted local trigonometric transforms. And we classify the decomposed signal by M uniform subbands to each subinterval. The energy of each subband is used as speech features. This feature is applied to VQ and HMM.