A hidden Markov-model-based trainable speech synthesizer (1999)

by Robert E Donovan, P C Woodland
Venue:Computer Speech and Language