Phoneme to Grapheme converter (P2G) between roman and hangul script for LING83600. Takes a text containing romanization of Korean lemms and produces list of likely equivalents in Hangul script.
To use: [python3 (shebang added but untested due to dev issues)] hangulize.py [--Syllable] source [output (hangul.txt by default)]
Note: --Syllable flag increases range of outputs and increases accuracy at possible cost to run-time performance.
Required:
- python 3.6.5 or later.
- pynini package and associated dependencies (please see requirements at http://www.openfst.org/twiki/bin/view/GRM/Pynini)
- jamotools package: pip install jamotools
Common Issues:
- Software assumes that provided forms are romanizations of Hangul graphemes. As such, assumes syllable composition to be VALID PHONETIC APPROXIMATIONS OF KOREAN WORDS. Usage of English exclusive phonemes (/f/, /v/, /z/) will result in ERROR token in output text string. -Above issue also applies to mophological restrictions in Korean. (e.g. conversion of "string"). Will result in runtime error. (Future work intends to alleviate this issue.)
Thanks to:
-Pynini:
K. Gorman. 2016. [Pynini: A Python library for weighted finite-state grammar compilation]
(http://openfst.cs.nyu.edu/twik/pub/GRM/Pynini/pynini-paper.pdf).
In Proc. ACL Workshop on Statistical NLP and Weighted Automata, 75-80.
-Jamotools
-Wiktionary for Korean Lemmas and associated romanizations.
-Universal Dependencies Treebank for Training Data.
-OpenFst group: Cyril Allauzen, Michael Riley, Johan Schalkwyk, Wojciech Skut and Mehryar Mohri, "OpenFst: A General and Efficient Weighted Finite-State Transducer Library", Proceedings of the Ninth International Conference on Implementation and Application of Automata, (CIAA 2007), volume 4783 of Lecture Notes in Computer Science, pages 11-23. Springer, 2007. http://www.openfst.org.