<?xml version="1.0" encoding="UTF-8"?> <article key="pdf/4756" mdate="2008-02-22 00:00:00"> <author>Erika Pigliapoco and Valerio Freschi and Alessandro Bogliolo</author> <title>Inferring Hierarchical Pronunciation Rules from a Phonetic Dictionary</title> <pages>274 - 281</pages> <year>2008</year> <volume>2</volume> <number>2</number> <journal>International Journal of Computer and Information Engineering</journal> <ee>https://publications.waset.org/pdf/4756</ee> <url>https://publications.waset.org/vol/14</url> <publisher>World Academy of Science, Engineering and Technology</publisher> <abstract>This work presents a new phonetic transcription system based on a tree of hierarchical pronunciation rules expressed as contextspecific graphemephoneme correspondences. The tree is automatically inferred from a phonetic dictionary by incrementally analyzing deeper context levels, eventually representing a minimum set of exhaustive rules that pronounce without errors all the words in the training dictionary and that can be applied to outofvocabulary words. The proposed approach improves upon existing ruletreebased techniques in that it makes use of graphemes, rather than letters, as elementary orthographic units. A new linear algorithm for the segmentation of a word in graphemes is introduced to enable outof vocabulary graphemebased phonetic transcription. Exhaustive rule trees provide a canonical representation of the pronunciation rules of a language that can be used not only to pronounce outofvocabulary words, but also to analyze and compare the pronunciation rules inferred from different dictionaries. The proposed approach has been implemented in C and tested on Oxford British English and Basic English. Experimental results show that graphemebased rule trees represent phonetically sound rules and provide better performance than letterbased rule trees. </abstract> <index>Open Science Index 14, 2008</index> </article>