When using ROL feature, I prefer to use the select tool to read certain paragraph instead of letting it read through the entire page. For some files, ROL would read only read a single word or portion of a word when I click the paragraph with the select tool. I sort of figured out the problem may be coming from the file not having been tagged. Tagging it sometimes solves the problem but not consistently. Anyone has a solution to this problem? Thanks.
read out loud problem
2010-01-22 21:59:20
My Product Information:
Acrobat Pro 9.0, Windows
Acrobat Pro 9.0, Windows
2010-01-23 18:41:29
#2
Thank you for the response. But I do not think my problem is as you described. The PDF files with problems are from medical journals and are of good quality. The ROL has no problem reading the entire page or to the end of the paper, but when I tried to have it read only a single paragraph it sometimes would read only part of a word. For example, the word "paragraph" would be read as "para".
2010-01-23 19:07:01
#3
Well, that can be awkward.
I do like ROL's basic functionality; but, the operative word is "basic" .
As to the source PDF, if the page content is a scanned image of text, it is still is worth a Save As to *.txt for a look-see.
Having spent more hours than I care to remember processing PDFs containing
scanned text for OCR via Acrobat, Capture Cluster, and AdLib I've learned that
even 'quality OCR' output from 'quality' source paper through industrial strength
scanners that are routinely cleaned and calibrated can hold surprises.
Regardless, I suspect it is more an issue of ROL's "basic" nature.
Be well...
I do like ROL's basic functionality; but, the operative word is "basic" .
As to the source PDF, if the page content is a scanned image of text, it is still is worth a Save As to *.txt for a look-see.
Having spent more hours than I care to remember processing PDFs containing
scanned text for OCR via Acrobat, Capture Cluster, and AdLib I've learned that
even 'quality OCR' output from 'quality' source paper through industrial strength
scanners that are routinely cleaned and calibrated can hold surprises.
Regardless, I suspect it is more an issue of ROL's "basic" nature.
Be well...
2010-01-23 19:22:28
#4
I tried saving as txt file and it looked fine. The problem persisted. Looks like I'll have to live with the problem. Thanks for your help regardless.
Yes, When asking ROL to parse a single line it will stop at some characters in a line.
Some examples of what I've encountered are:
--| a nonbreaking space
--| a discretionary hyphen
Such characters, in a PDF, would have come from the authoring file where the content author inserted them.
Typically, not a problem when you have ROL parse the entire page.
For a scanned image of text, there'd have to be OCR applied. If Searchable Image or Searchable Image (Exact)
are used and depending on the quality of the source paper's imprint, the OCR'd text can contain some funky stuff at times.
These can be a speed hump for ROL parsing a single line.
If using either of the OCR methods above, a quick check is to save as to a text file after OCR and Save of the PDF.
The text file's content is highly indicative of what you'll get from ROL (with "funky characters being potential speed humps).
Be well...
Be well...