Google Translate is adding 110 new languages to its library, the largest expansion ever on the platform. The update uses Google’s major language model PaLM 2, an artificial intelligence tool that helps accurately translate into a wider range of languages than before. These languages are spoken by approximately 614 million people, or about 8% of the world’s population.
The list includes some commonly spoken languages, dialects and languages indigenous to smaller communities. Notably, African languages saw their greatest expansion, with Fon, Kikongo, Luo, Ga, Swati, Venda and Wolof joining the list. At the other end of the spectrum, Cantonese is probably one of the most spoken languages on the new list, along with Punjabi (Shahmukhi), the most spoken language in Pakistan.
There is also the Sicilian version of Italian, Manx, a Celtic language spoken on the Isle of Man that is almost extinct, and a creole language from Papua New Guinea called Tok Pisin.
PaLM 2 Talk
For languages that combine regional dialects and different spelling standards, Google chooses what can be best understood by most people, such as the Romani language offered by Google Translate, which includes three different dialects.
The PaLM 2 LLM made the update possible, increasing Google Translate’s ability to learn and switch between languages efficiently. This model is particularly adept at handling languages that are closely related, such as Awadhi and Marwadi, to Hindi or the various French Creoles. PaLM 2’s advanced capabilities allow it to manage the nuances and variations within these languages, delivering more accurate and culturally relevant translations.
The application of PaLM 2 to Google Translate is also interesting because of its origins as a tool to facilitate communication between humans and AI. For example, both PaLM and PaLM 2 have been used to help robots perform tasks and process human commands into steps to complete those tasks.
Perhaps the best part is that it’s available on the web or through the Google Translate app on Android and iOS.