Microsoft's Dataset For Machine Translation, NTREX, aims to bridge the language divide with 128 languages, each having 2000 sentences.
