Unlocking Zero-Resource Machine Translation to Support New Languages in Google Translate

Google Translate is using few shot translation to support low resource languages without as much data available.

Relying on the benefits of transfer learning in massively multilingual models, we train a single giant translation model on all available data for over 1000 languages. The model trains on monolingual text for all 1138 languages and on parallel text for a subset of 112 of the higher-resourced languages.

1 Like