Exploring automatic similarity measures for unit selection tuning
Daniel Tihelka, Jan Romportl · 2009
The present paper focuses on the current handling of target fea-tures in the unit selection approach basically requiring huge cor-pora. In the paper there are outlined possible solutions based on measuring (dis)similarity among prosodic patterns. As the start of research, the feasibility of (dis)similarity estimation is ex-amined on several intuitively chosen measures of acoustic sig-nal which are correlated to perceived similarity obtained from a large-scale listening test. Index Terms: speech synthesis, unit selection, target features, prosodic patterns, perceived similarity, signal similarity, multi-