Abstract | ||
---|---|---|
A finite-state method, based on leftmost longestmatch 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 rule and a 'lazy' variant of Brill's algoritm), trained on only 40K words, reaches 99%. |
Year | Venue | Keywords |
---|---|---|
2000 | north american chapter of the association for computational linguistics | finite-state method,data-oriented method,small set,transformation-based learning,large rule,hand-crafted conversion rule,phoneme accuracy,leftmost longestmatch replacement,finite state,best system |
DocType | Volume | ISSN |
Conference | cs.CL/0003074 | Proceedings of NAACL-2000, Seattle, WA |
Citations | PageRank | References |
5 | 0.64 | 12 |
Authors | ||
1 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gosse Bouma | 1 | 483 | 70.88 |