A Finite State and Data-Oriented Method for Grapheme to Phoneme Conversion

Simple item page
dc.creatorBouma, Gosse
dc.date2000-03-23
dc.date.accessioned2026-07-07T03:16:07Z
dc.date.available2026-07-07T03:16:07Z
dc.descriptionA finite-state method, based on leftmost longest-match replacement, is presented for segmenting words into graphemes, and for converting graphemes into phonemes. A small set of hand-crafted conversion rules for Dutch achieves a phoneme accuracy of over 93%. The accuracy of the system is further improved by using transformation-based learning. The phoneme accuracy of the best system (using a large set of rule templates and a `lazy' variant of Brill's algoritm), trained on only 40K words, reaches 99% accuracy.
dc.description8 pages
dc.identifierhttps://arxiv.org/abs/cs/0003074
dc.identifierhttp://arxiv.org/abs/cs/0003074
dc.identifierProceedings of NAACL-2000, Seattle, WA
dc.identifier.urihttp://salesiana.dossiersoluciones.com/handle/123456789/30232
dc.subjectComputation and Language
dc.subjectI.2.7
dc.titleA Finite State and Data-Oriented Method for Grapheme to Phoneme Conversion
dc.typetext

Files

Collections

arXiv