Hi,
I exploded the hindi training data and looked through hin,unicharset, and it seems to be missing at least the anubanda "ऽ" and full-stop "॥" characters. Will the box/tiff files be made available? I'm interested in adapting the hindi data for sanskrit texts in devanagari.
Thanks