Karnevskaya, E. Current Problems of Prosodic Modelling for Speech Synthesis. baj 2005, 21-25.