Previous work has demonstrated that multilayer perceptrons may be used to transcribe orthographic text into a phonemic representation. It is shown that this technique may be applied to a large dictionary of over 70,000 words without reduction in performance. The dictionary was divided into sub-dictionaries containing regularly and irregularly pronounced words using a rule-based system. The multilayer perceptron performed better with the regularly pronounced words. However, combining two separately trained networks resulted in little better performance than a single network trained on the whole dictionary.
Cite as: Ainsworth, W.A., Pell, B. (1989) Connectionist architectures for a text-to-speech system. Proc. First European Conference on Speech Communication and Technology (Eurospeech 1989), 1125-1128, doi: 10.21437/Eurospeech.1989-43
@inproceedings{ainsworth89_eurospeech,
author={William A. Ainsworth and B. Pell},
title={{Connectionist architectures for a text-to-speech system}},
year=1989,
booktitle={Proc. First European Conference on Speech Communication and Technology (Eurospeech 1989)},
pages={1125--1128},
doi={10.21437/Eurospeech.1989-43}
}