A hybrid speech recognition system using HMMs with an LVQ-trained codebook.

Hitoshi Iwamida, Shigeru Katagiri, Erik McDermott, Yoh'ichi Tohkura

Browse the full ICASSP paper archive.