My system for doing OCR is now up and running. This is Window-Eyes
1.02 with a Sounding Board, an HP III-P scanner with Accupage, and
WordScan 4.0 Plus. The advantage of WordScan is that it has a
built-in proofing editor. Window-Eyes can find its caret, and you
move from one suspicious character or word to the next with the tab
key. This editor doesn't come up until you have OCRed the document
and closed it, so you may have to do a trial run on the first page,
then scan multiple pages. The manual says that there is a
conflict between the zoning done by Accupage and the
auto-orientation feature, so I haven't really tested that
yet. It's certainly not as speech-friendly as a dedicated
system, but adheres to Windows conventions and should work
with a wide range of screen readers.
Accupage, Wordscan and the Proofing Editor take up 9 megs of RAM when
all are running (according to the display you get in Help About inside
Program Manager). So I would not reccommend this system with less
than 16 megs of RAM (a permanent swap file will make it pretty slow).
I don't have a good way to compare accuracy against other products,
but it is working on many of the things I have tried, including a
couple of pages of "Soap Opera Digest" which used to bring the
Kurzweil machine to its knees.
Lloyd Rasmussen
Senior Staff Engineer
National Library Service f/t Blind and Physically Handicapped
Library of Congress 202-707-0535
lras@loc.gov
This archive was generated by hypermail 2b29 : Sun Dec 02 2012 - 01:30:04 PST