STATISTICAL METHODS FOR AUTOMATIC PROSODIC BREAK DETECTION IN A TEXT-TO-SPEECH SYSTEM
«Speech Technologies Center», Ltd.; scientific researcher
O. G. Khomitsevich
STC Ltd., St. Petersburg; Senior Scientist
S. V. Rybin
STC Ltd., St. Petersburg; ITMO University, Department of Speech Information Systems; Leading Programmer; Associate Professor
Abstract. Application of statistical methods for predicting positions and durations of prosodic breaks in a text-to-speech system is proposed. The methods are shown to ensure better results as compared with a baseline rule-based system.
Keywords: prosodic breaks, prosodic boundaries, pauses, speech synthesis, statistical models