SPEECH SYNTHESIS BY RULE USING AN OPTIMAL SELECTION OF NON-UNIFORM SYNTHESIS UNITS.

A synthesis scheme is proposed for the optimal selection and extraction of synthesis units. This synthesis scheme has two advantages in the usage of speech segment units. One is the flexible use of nonuniform synthesis units and the other is the optimal choice of a unit sequence for an input phoneme string using appropriateness measures. In this synthesis system, all phoneme subsequences in each synthesis unit are also used as synthesis units. To facilitate the search for appropriate unit candidates, a phoneme sequence entry dictionary is constrictive for a given speech set. Statistical analysis results are presented for Japanese phoneme sequences based on distributions in a word dictionary and a body of text. This investigation provides an estimate of the appropriate number of Japanese phoneme sequences in this synthesis scheme.