In the Partnership for Applied Biblical NLP, we are defining a “shared task” for word segmentation - basically, a contest to see which software can do it best. The goal is to create NLP software that can do it better. We are considering using the format Paratext uses in Wordanalyses.xml, and we need high quality data for a variety of languages.
I have these files for some languages where I am at least an observer on the project, but I need more, particularly for languages where a lot of work has been done on the morphology in either the Interlinearizer or the Wordlist.
We are interested in both languages of wider communication and languages that are morphologically complex and represent different kinds of languages. If you have data you are willing to share, please let me know!
[Email Removed]
