HMM representation of quantized articulatory features for recognition of highly confusable words.

Kevin Erler, Li Deng

Browse the full ICASSP paper archive.