This is an attempt to experiment with substituting the Henry Ornithography with something that will only make use of standard ASCII characters. This will make it easier to feed into machine learning models with off the shelf code. Where possible, the scheme matches the Cayuga keyboard map.
See Sge:no
Scheme consists of
the e-ogonik ==> v
the o-ogonik ==> c
glottal ==> ]
stress marker will be the vowel capitalized
the elongated vowel sound : ==> double vowel
breathless vowel is eliminated as it was just going to be too much of a hassle. It could be supported with an underscore but this decision will make identification require less examples and conforms to upper Cayuga grammar.
The modified words look pretty horrible but all front facing words will be modified to the Henry structure anyway.
I can assure you that machine learning models have been able to handle exotic languages with non-ASCII characters such as French or German for some time now. It's more important to have a uniform orthography than to fit it into ASCII, because otherwise it's hard to mix text from different sources. Besides, sentences on Tatoeba can be front-facing anyway. So if this sentence is usually written "Sgę́:nǫˀ," I think you should write it that way as well.
Thank you for your feedback
Logs
added by robot_fury, January 27, 2019
license chosen by robot_fury, January 27, 2019
deleted by robot_fury, May 4, 2019