Parallel Corpora for Machine Translation

Parallel corpora for nearly 400 language pairs and numerous multilingual combinations, including 10 million quality segments. 

The segments consist of corpus-derived, manually curated full sentences and short phrase, including their translation equivalents. They are based on dictionary examples of usage that have been created over the years by expert linguists, lexicographers and translators worldwide. 

The data can be applied to enhance the training of language learning models to boost the performance of Neural Machine Translation engines.

The languages include: Arabic, Chinese (Simplified), Danish, Dutch, English, Finnish, French, German, Greek, Hebrew, Italian, Japanese, Korean, Latin, North Sami, Norwegian, Polish, Portuguese (Brazil / Portugal), Russian, Spanish, Swedish, and Turkish.

In addition to general language vocabulary, there are segments for more than one hundred vertical domains.