Corpora used to train Punkt Segmenter in German

[AugustinErnoult]

Hi,
Thank you very much for your amazing work.
I used NLTK to segment a german text. I see that this language is available and the sentence tokenizer gives quite good result with the default training. However, I would like to now on which corpora it as been trained and I don't find any german text in this list (or I missed it) .
Do you know where I can find an explicit list of text used?
Thank you

[AugustinErnoult]

I found it in "nltk-data/punkt/tokenizer/README". For German it seems to be only the corpus initially used by Strunk and Kiss.