This contribution gives an overview of the performance evaluation of our text-to-speech conversion system for Spanish. This rule-based system comprises both a linguistic and a phonetic component. The performance test involved the linguistic output as well as the phonetic output. The linguistic output (which shows the workings of rules concerning grapheme-to-phoneme conversion, assimilation, word stress, syllabification, and intonation) was checked on the basis of a small text corpus. The phonetic output, i.e. the synthesized speech, was tested with respect to intelligibility and naturalness, using both an expert and naive listeners.
Cite as: Gerwen, R.P.M.W.v., Vieregge, W.H., Kerkhof, M.P.A.M. (1989) Evaluation of an automatic text-to-speech conversion system for Spanish. Proc. Speech Input/Output Assessment and Speech Databases, Vol.2, 107-110
@inproceedings{gerwen89_sioa,
author={R. P. M. W. van Gerwen and Wilhelm H. Vieregge and M. P. A. M. Kerkhof},
title={{Evaluation of an automatic text-to-speech conversion system for Spanish}},
year=1989,
booktitle={Proc. Speech Input/Output Assessment and Speech Databases},
pages={Vol.2, 107-110}
}