@drammock -- as discussed a long time ago, this PR merges the various "sources" (initial UW "ph" inventories; work with Dr. Green "gm", and work at UZH "uz") into a single "ph"oible source, i.e. a convenience sample aimed at filling typological gaps, and aimed at comprehensive phonemes, allophones, and tones extraction.
I left the files as are for historical reasons (e.g. if we identify a systematic mistake in one of inventory data files, they have slightly different input formats (some need gap filling)), but moved them all into the PH folder, updated the aggregation script, and regen'd the data.
@drammock -- as discussed a long time ago, this PR merges the various "sources" (initial UW "ph" inventories; work with Dr. Green "gm", and work at UZH "uz") into a single "ph"oible source, i.e. a convenience sample aimed at filling typological gaps, and aimed at comprehensive phonemes, allophones, and tones extraction.
I left the files as are for historical reasons (e.g. if we identify a systematic mistake in one of inventory data files, they have slightly different input formats (some need gap filling)), but moved them all into the
PHfolder, updated the aggregation script, and regen'd the data.And updated the README.
Let me know if you want any changes.