Translation-Help

OCR for scanned documents in PDF format
- Linux has a variety of tools to extract text from a picture (e.g. scanned document) and convert it to text suitable for text processing tools (OCR). The small script below shows the basic steps to convert a PDF document containing a scanned document in image format. Tools used (available as packages in most Linux distributions): pdftk, ImageMagick, tesseract. The example assumes that images do not need to be rotated. Please note: the delete command at the end is not suitable for every situation!
- ```
if [ $# == 0 ]; then
   echo "usage: ocr.sh pdf-file-name"
   exit
fi;
if [ ! -f $1 ]; then
  echo "$1 does not exist"
  exit
fi;
pdftk "$1" burst dont_ask
  for f in pg_*.pdf
do
  echo "Convert PDF to TIFF for file: $f ..."
  convert -depth 8 -density 300 -quality 100 \
     "$f" "$f.tiff"
  rm -f "$f"
done
for f in pg_*.tiff
do
  echo "OCR recognition for $f ..."
  tesseract "$f" "$f.txt"
done
cat pg_*.txt > result.txt
rm pg_*
```
- The converted text will be placed in "result.txt".

Converting TTX to CSV or TMX with tikal (Okapi)
- Okapi-Tools (Okapi-Framework) are cross-platform and open-source tools for a variety of useful features for translation and localisation. This small example shows, how to convert a Trados TTX-file into a CSV-file or into a TMX translation memory using the command line tool Tikal.
- You can find the tools and the installation instructions under the link above. Once you have the tools installed, you can easily convert the format of any TTX-file using the following commands:
- Convert TTX to CSV: ${PATH-TO-TOOLS}/tikal.sh -2tbl x.ttx -csv
- This will create a file called x.ttx.txt.
- Convert TTX to TMX: ${PATH-TO-TOOLS}/tikal.sh -2tmx x.ttx
- This will create a file called x.ttx.tmx

Combining files with Ghostscript
- Most of the time Ghostscript is already installed on the system for other reasons. Should it be missing, you can install it using the package maintenance system of your distribution.
- For example, on Fedora (RPM-based):
- yum install ghostscript
- After the installation use the following command to combine PDF-files:
- gs -dNOPAUSE -sDEVICE=pdfwrite -sOUTPUTFILE=out.pdf -dBATCH pdf1.pdf pdf2.pdf
Combining files with PDFtk
- Most of the time PDFtk has to be installed on the system. This is accomplished most easily using the package maintenance system of your distribution.
- For example, on Fedora (RPM-based):
- yum install pdftk
- After the installation use the following command to combine PDF-files:
- pdftk pdf1.pdf pdf2.pdf cat output out.pdf
Splitting PDF-files with PDFtk
- Most of the time PDFtk has to be installed on the system. This is accomplished most easily using the package maintenance system of your distribution.
- For example, on Fedora (RPM-based):
- yum install pdftk
- After the installation use the following command to split PDF-files
- pdftk pdf1.pdf burst
- This splits the PDF-file into individual pages named pg_0001.pdf, pg_0001.pdf etc. The parameters output allows you to name the output files differently.