OCR: Problem with german "umlaute"

DEVONthink Tips
1

When I convert a scanned document into a searchable PDF, the typical german characters are not properly recognized. For example, the umlauts ‘ä’, ‘ö’ and ‘ü’ all lose their dots, and the letter ‘ß’ becomes ‘B’. This seems to be a systematic error, because I cannot find those characters in any document I tested, and apart from this minor problem, the text recognition is really impressive.

I tried using OCR with different qualities of the source document, up to 600dpi, without success. In the OCR preferences, recognition is set to ‘accurate’, and the primary language is german.

I use version 2.0pb7 of DT Pro Office.
Any hints?

  • Markus
2

Make German the only language (or remove French) for now until public beta 8 is out.

3

Thanks a lot, that did the trick!

  • Markus