1

I have a few books in printed format. They were translated from English to Norwegian before computers were mainstream, so I ran the translation through a scanner and OCR-ed it. The English version was already digitized.

There are OCR errors that I need to correct, and I also need to match the paragraphs in the Norwegian translation to the English original.

It seems I should use some Translation Memory program like Zanata as the tool and repository for the translations, but before I can do that I need to create a TMX file that matches the segments between the existing translations

Any idea how I can go about doing that? Or am I attacking this problem at the wrong angle?

3

1 Answer 1

0

If you search for "Align Translations", you will find a few tools that are designet specifically for aligning paragraphs (or sentences) in already translated documents.

I used sourceforge.net/projects/aligner, which is cross plattform. I am not overly impressed by the tool- but it gets the job done, as far as the question above goes (it allows you to align text and then produces a TMX-file)

However, I ended up leaving the tool for a reason that is not clear from the question above. I needed the aligner to output a text file that matched the original exactly, with regards to number of line breaks. Even though LF Aligner stores an ongoing project in text files, there were a few issues:

  • When I opened up a project a second time, some of the aligning was messed up. Specifically this pertained to places I added an empty paragraph.
  • General problems handling empty paragraphs in both original language and translation.

I will add more info to this post if I find a tool that is able to do this.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .