On 12 Mar 2021, at 8:58 PM, James Benstead <james.benstead@...> wrote:



Thank you all! That's some really helpful information regarding OCR. I had been using the OCR functions in Acrobat Pro, but I'll give what you suggest a run through, too.

I'm also interested in more general guides to PDF accessibility. I've got various PDFs here that don't need OCR done to them -- they are there on the screen as actual text -- but when I go to read them I still get stuck. Some won't read past the first page, for example; others aren't tagged (this is almost all of them!) and while I've had some success with the AutoTag feature in Acrobat Pro this certainly hasn't solved all of my problems. What I'm really looking for, I suppose, is a list of all the tricks people have developed to deal with PDFs that don't need OCR work, but still aren't accessible.

Jim

--

Too brief? Here's why! http://emailcharter.org

On Thu, 11 Mar 2021 at 22:57, Brian Vogel <britechguy@...> wrote:

Scott,

            I really do not know how it differs from JAWS convenient OCR.  What I do know is that PDF X-Change Viewer has been an incredibly good, and incredibly accurate, OCR scanner that has worked very well on some pretty rough originals.  It also supports many additional language packs, at no cost, beyond English, French, German, and Spanish (if I'm remembering the four correctly) that come with the native utility.  I had a client who was a translator from Swedish to English and she used it with the Swedish language pack for documents that were in Swedish.  It does not, however, do any automatic language switching.  It presumes that a given document will be in a single language, so it wouldn't be good for, say OCRing an image scanned page from a language text where both languages are used interspersed on the same page.

             What I liked about it was that it was easy both to run the OCR and to save the resulting file such that you'd never have to do OCR again and could send it to others with the text layer already there.  I had my client who was a grad student send these to his instructors, and asking them to replace the existing image scanned PDFs with those with the OCR text layer so that later students who were blind or visually impaired would not have to do any OCR processing on them.
--

Brian - Windows 10 Pro, 64-Bit, Version 20H2, Build 19042  

To think is to differ.
      ~ Clarence Darrow