Google Translate has expanded its language support to more than 500 million people, including languages such as Cantonese, Manx, and NKo. This represents around 8% of the world’s population, with a focus on African languages in this latest update. Some of the new languages come from small indigenous communities, while others have active revitalization efforts.
Afrikaans, spoken in Djibouti, Eritrea, and Ethiopia, had the most voluntary community contributions in this release. Cantonese, a highly requested language, was challenging to train models for due to its overlap with Mandarin in writing. Manx, a Celtic language, almost became extinct but has seen a revival effort, resulting in thousands of speakers on the Isle of Man.
NKo, a standardized form of Mandinka languages in West Africa, has a unique alphabet and an active research community. Punjabi (Shahmukhi) and Tamazight (Amazigh) are also among the newly supported languages, representing diverse linguistic communities.
Tok Pisin, a creole of English origin and the lingua franca of Papua New Guinea, is another language now supported by Google Translate. The platform considers various linguistic varieties and includes multiple dialects when selecting which forms of a language to support. PaLM 2 technology has assisted in efficiently learning closely related languages, allowing Translate to expand its offerings over time.
As technology continues to advance and Google partners with experts and native speakers, more languages and dialects will be supported in the future. Users can access translations in these newly supported languages by visiting the Help Center or using the Google Translate app on Android and iOS.
Article Source
https://blog.google/products/translate/google-translate-new-languages-2024/