Save and apply common regex-fixes before handing over the digitized text for proof-reading

[vvasuki]

Some OCR tools result in predictable errors, which can be fixed by regex replacement, as pointed out in this thread. These regexes should be saved in a bash script, which may be of the following form:

perl -pi -e 's/FINDTEXT/REPLACETEXT/' someFileName

This script should then be used on the OCR output before passing it on the proof reader.