These are two non-optimized, quick and dirty, algorithms for syllabifying Romanian words written orthographically (and not in IPA). They were written in order to have a baseline accuracy for the machine learning experiments described in Liviu P. Dinu, Vlad Niculae, Octavia-Maria Sulea. Romanian Syllabication Using Machine Learning. TSD 2013: 450-456 and implemented here
ortho_syllabify.py
- follows the end-of-the-line hyphenation rules listed in DOOM
- can be quickly tested here: http://www.codeskulptor.org/#user5-Sx4RXVO6ROjUMze-7.py
syllabify.py
- follows MOP and Ioana Chitoran's constraint based approach to Romanian Phonology
- can be quickly tested here: http://www.codeskulptor.org/#user5-AN4aaEbwtq49Ar4-18.py