Parallel corpora is a set of parallel texts in the domain of Law and Health, with 1 G per language. Languages: cs-pt, de-pt, en-pt, es-pt, fr-pt, it-pt, and pt-sk.