Arabic-L:GEN:OCR program responses

[Dilworth Parkinson]

------------------------------------------------------------------------
Arabic-L: Wed 10 Oct 2012
Moderator: Dilworth Parkinson <dilworth_parkinson at byu.edu>
[To post messages to the list, send them to arabic-l at byu.edu]
[To unsubscribe, send message from same address you subscribed from to
listserv at byu.edu with first line reading:
           unsubscribe arabic-l                                      ]

-------------------------Directory------------------------------------

1) Subject:OCR program response
2) Subject:OCR program response

-------------------------Messages-----------------------------------
1)
Date: 10 Oct 2012
From:Karen McNeil <karenlmcneil at gmail.com>
Subject:OCR program response

Regarding OCR programs:

I was also looking for an OCR program for my corpus work, and I found
that none of the ones I tried were able to deal at all well with my
materials.  (Which were good-quality printed books, but in dialect.)

I found that it was much more effective to use workers on Amazon Turk
for this kind of work.  I put up the material I needed transcribed,
with each page as a separate 'hit' which paid $0.20.  The work was
done very quickly (all the hits had been completed in a few hours),
and the batches that I accepted had almost perfect accuracy.

I don't know how many pages you have to do, and so if this would be
cost prohibitive at even such a low rate, but I had several hundred
pages transcribed this way with great success.

Good luck,
Karen McNeil

--------------------------------------------------------------------------
2)
Date: 10 Oct 2012
From:Stewart Felker <stewart.felker at gmail.com>
Subject:OCR program response

May I ask which programs you've tried so far? I've heard Sakhr is quite
good for Arabic; but I haven't been able to run it on my computer yet. I
know VERUS is also supposed to be excellent - but it's like $1000. I've
heard OmniPage is a good, cheaper alternative - but, again, haven't been
able to test it.

--------------------------------------------------------------------------
End of Arabic-L: 10 Oct 2012