Inductive Learning of Pronunciation Rules by hypothesis Testing and Correction
Proceedings of the Seventh International Joint Conference on Artificial Intelligence. Learning. Pages 109-113.
Abstract
This paper describes a system that learns the rule s of pronunciation inductively. It begins with a set of 26 rules for single-letter pronun ciation. Individual words are presented to it, and the s ystem uses its rule set to hypothesise a pron unciation. This is compared with a dictio nary pronunciation, and if any part of the pronunciation is incorrect new rules are created to hand le the word as an exception condition. These rules are checked for similarity with othe r s already produced, and where suitable a "general" rule is produced to deal with two or more created rules. The effect is to produce rules that are more and more general, and these approach the ge neral pronunciation rule sets that have been produc ed manually by other workers.
@inproceedings{ijcai1981p24,
title = {Inductive Learning of Pronunciation Rules by hypothesis Testing and Correction},
author = {Oakey, S. and Cawthorn, R. C.},
booktitle = {Proceedings of the Seventh International Joint Conference on
Artificial Intelligence, {IJCAI-81}},
publisher = {International Joint Conferences on Artificial Intelligence},
editor = {Roger C. Schank},
pages = {109--113},
year = {1981},
month = {8},
note = {Learning},
url = {https://ijcai.org/proceedings/1981/24},
}
TY - CPAPER TI - Inductive Learning of Pronunciation Rules by hypothesis Testing and Correction AU - Oakey, S. AU - Cawthorn, R. C. T2 - Proceedings of the Seventh International Joint Conference on Artificial Intelligence, IJCAI-81 PB - International Joint Conferences on Artificial Intelligence PY - 1981 SP - 109 EP - 113 UR - https://ijcai.org/proceedings/1981/24 ER -Copied
Machine-readable versions of this paper, for tools and AI agents:
