The system presented in this paper produces synthetic speech by concatenating three types of units to generate a syllable. While the initial unit is a classic demisyllable, the final demisyllable is split into a VC diphone and an optional suffix containing the final consonant cluster. Rules controlling the synthetisizing process are formulated in a language especially developed for text-to-speech synthesis. A set of rules for duration control and concatenation is presented.
Cite as: Portele, T., Sendlmeier, W., Hess, W. (1990) Hadifix : a system for German speech synthesis based on demisyllables, diphones and suffixes. Proc. First ESCA Workshop on Speech Synthesis (SSW 1), 161-164
@inproceedings{portele90_ssw,
author={Thomas Portele and Walter Sendlmeier and Wolfgang Hess},
title={{Hadifix : a system for German speech synthesis based on demisyllables, diphones and suffixes}},
year=1990,
booktitle={Proc. First ESCA Workshop on Speech Synthesis (SSW 1)},
pages={161--164}
}