On Saturday, October 25, 2014 10:11:14 PM UTC-3, xorxes wrote:
I have changed the stress rule, which I think is what's causing the problem, to:
stress <- consonant* h? y? syllable pause
I re-evaluated the camxes test corpus and the current contents of jbovlaste with this modification, and there are no changes beyond what has been discussed in the other thread for v122 of the morphology.
mi'e la mukti mu'o