Some conversation was going on regarding the issue in http://egroups.com/group/bdosdn/
Thanks Omi On 1/30/06, Sankarshan Mukhopadhyay <[EMAIL PROTECTED]> wrote: > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > > > - -------- Original Message -------- > Subject: [silk] What's the state-of-the-art for Bengali OCR > Date: Sun, 29 Jan 2006 09:42:47 +0530 > From: Sujai <[EMAIL PROTECTED]> > Reply-To: silklist@lists.hserus.net > To: Silkist <silklist@lists.hserus.net> > > Hi all. I figured Silklist would be the best place to start asking this: > > My aunt helps with the Special Ed Cell at Jadavpur University > (Kolkata) and the Cell has so far been providing assistive and > accessible technologies and services like screen readers, braille > printers, etc. > > They are planning to buy a more elaborate document/image scanner than > the simple SOHO flat-bed they have right now, and hope to also make > electronic versions of Bengali texts (that are in the public domain) > available on their website. > > The question is - how easy is it to get from Bengali printed text to > electronic text? Searches so far have revealed that the Indian > Statistical Institure (ISI, Kolkata) and CDAC, Pune have a Bengali OCR > package but the sites are singularly unhelpful when it comes to > details about how it works, features, and accuracy. > > If any of you have any information on what the best way to go about > this would be (suggestions for other packages, pointers to people), > that would be great. > > Thanks > > - -- > Sujai > www.ylog.org/sujai
