Hi all,

The client has given us existing translations of their string content in TXT format. They would like us to build TMs for them from those existing strings. We've explained the risks associated with this approach (being that we don't know who provided that original content).

For Chinese Traditional & Simplified, Korean, Arabic, PT-PT and Russian, we have corrupted text in the TXT file. When we run that file in Trados, Trados displays the content correctly. I can export a bilingual review document which also displays the content correctly, though this is breaking the target at a different spot than the original. This causes problems for alignment. I tried copy source to target in the Trados environment, and exporting a TXT file, and that displays the font corruption once again.

Also, the source and targets they provided us are not exact matches, so we're having to sort the content by coding identifier in order to even align. We're doing that in an Excel file, and we can get the English and target content to line up (with holes for any untranslated text). However, the text shows up as corrupted, and the solutions above have not worked to overcome this.

Any ideas for getting a TM out of these files?

Thank you,

Alaina