First European Conference on Speech Communication and Technology

Paris, France
September 27-29, 1989

Connectionist Architectures for a Text-to-Speech System

William A. Ainsworth, B. Pell

Department of Communication and Neuroscience, University of Keele, UK

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.

Full Paper

Bibliographic reference.  Ainsworth, William A. / Pell, B. (1989): "Connectionist architectures for a text-to-speech system", In EUROSPEECH-1989, 1125-1128.