Фонологическое моделирование

SRI has developed a speaker-independent con- tinuous speech, large vocabulary speech recogni- tion system, DECIPHER, that provides state-of-the- art performance on the DARPA standard speaker- independent resource management training and …...