Software for scanning of Indian rarities?

[Claude Setzer]

I believe there is only one such software, that has passed through several
hands over the years (including David Hewlett of Hewlett Packard). I have
access to it, however, I have not actually used it because it is very slow,
runs only on an old MAC, and makes too many mistakes to actually save any
time (due to required editing). Even the best English ones, that require
extensive knowledge of English usage and dictionary to achieve good accuracy
are not all that good. Some of the best English ones only achieve 99%
accuracy on reasonably good text. The devanagari one is more like 95%, which
means a huge amount of editing.

Some work is being done on adding an extensive Sanskrit grammar and
dictionary to the above software, which could to bring it up to "English"
error levels, but it is on the back burner and certainly many years off.

Another option which may only a slight chance of being implemented is the
Verbot (PC base verbal robot) approach which carries on conversations in
English now and promises to be extended to other languages in the next few
years. I think these people are going to have better and less expensive
voice recognition than what is available now, and since Sanskritists  are
sometimes willing to put more attention on "consistent" pronunciation than
speakers of other languages, this might prove to be quite interesting as a
way to enter text.

Claude Setzer cssetzer at mum.edu

-----Original Message-----
From: Viktor V. Sukliyan <madhava at CH1.VSNL.NET.IN>
To: INDOLOGY at LISTSERV.LIV.AC.UK <INDOLOGY at LISTSERV.LIV.AC.UK>
Date: Sunday, March 15, 1998 10:46 AM
Subject: Software for scanning of Indian rarities?


>Does anybody out there know about availibility of software for
>OPTICAL CHARACTER RECOGNITION (OCR) of Indian fonts(Devanagari,
>Bengali etc) for purpose of scanning Indian texts with further
>conversion them in text files?
>It is known that we have this facility for european languages.
>What about Indian languages?
>
>With best regards
>Viktor Sukliyan.
>