Optimising selection of units from speech databases for concatenative synthesis

Concatenating units of natural speech is one method of speech synthesis 1 . Most such systems use an inventory of fixed length units, typically diphones or triphones with one instance of each type. An alternative is to use more varied, non-uniform units extracted from large speech databases containing multiple instances of each. The greater variability in such natural speech segments allows closer modeling of naturalness and differences in speaking styles, and eliminates the need for specially-recorded, single-use databases. However, with the greater variability comes the problem of how to select between the many instances of units in the database. This paper addresses that issue and presents a general method for unit selection. 1. INTRODUCTION The ATR -talk system for Japanese 4 efficiently selects non-uniform units from a large speech database but is specific to Japanese. English has more phonemes and more varying prosody, so a simple translation of the -talk system to English ...