+1 Anivar
On Mon, Jan 6, 2014 at 11:28 PM, Anivar Aravind <anivar.aravind(a)gmail.com>wrote;wrote:
On Mon, Jan 6, 2014 at 5:16 PM, Jayanta Nath <jayantanth(a)gmail.com> wrote:
In Indic languages , the basic issue is OCR. Till
date we have no OCR in
Indic languages. My opinion tying is not the solution, it can be temporary
solution.
There are many efforts on Training & Improving Tessearct for indian
languages . But there is no fully usable product yet . From a technology
point of view typing in wikisource helps in building training corpus for
OCR projects . Especially in languages like malayalam there are many script
variations against timeframe and Wikisource is a major effort that helps
to build a free licensed training corpus .
Anivar
_______________________________________________
Wikisource-l mailing list
Wikisource-l(a)lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/wikisource-l