Aligning a corpus with fast align

What you need

• A sentence-aligned parallel corpus
• Fast_align (get it here)
• Two apertium language packages

Process

First analyse the corpus with the language packages.

  cat 

Then remove superfluous tags (for example for lexical alignment, case is not really interesting).

  cat | sed 's/<\(nom\|acc\|gen\)>//g' > 

Create the input file for fast_align:

Run fast_align:

Symmetrise the alignments: