Thanks so far! I think I am on track now. I will generate my own csv's with a small script.
However my task is a bit more complex than just one file.
I have about 1000 clips, which will be english (e) and german (d) in high (h) and low (l) quality, which is designated by the file name. Each set of 4 files has an .xml that goes with it, which I will now make into an english and a german .csv.
All clips with number 619 contain the same "video"-content in high or low quality, just the speaker is German or English:
data00619-dh.mp4 <-- german high Q
data00619-dl.mp4 <-- german low Q
data00619-eh.mp4 <-- english high Q
data00619-el.mp4 <-- english low Q
data00619.xml <-- contains both english and german meta-information
data00619-d.csv <-- german
data00619-e.csv <-- english
I got this to work:
import text file (with filename of clip) into a dummy tag, e.g. comment
Source Values: %comment%
Guessing pattern: %title%;%album%;%genre%;%language%
Remove Fields: %comments%
Is there a more straightforward way to do this?
Thanks a bunch in advance!