07/10/2024
Gadgets & AppsMobileNewsfeed

Google Translate Expands with 110 New Languages Worldwide

Google Translate is breaking down language barriers to help people connect and understand the world better. With the latest technologies, we aim to make this tool accessible to more users. In 2022, we added 24 new languages using Zero-Shot Machine Translation, where a machine learning model learns to translate a language without prior examples. Additionally, we launched the 1,000 Languages Initiative to support the most spoken languages worldwide.

“We are now using AI to expand our language offerings. Our PaLM 2 large language model enables us to add 110 new languages to Google Translate, marking our largest expansion yet,” the company shared in a statement.

Supporting Over Half a Billion People

These new languages, from Cantonese to Qʼeqchiʼ, represent over 614 million speakers, covering about 8% of the global population. Some are major languages with over 100 million speakers, while others belong to small Indigenous communities or have few native speakers but are undergoing revitalization efforts. A significant portion of these additions, including languages like Fon, Kikongo, Luo, Ga, Swati, Venda, and Wolof, come from Africa, marking it its largest expansion of African languages.

Here are some newly supported languages:

  • Afar: Spoken in Djibouti, Eritrea, and Ethiopia, it had the most volunteer contributions in this launch.
  • Cantonese: Highly requested but challenging to support due to overlap with Mandarin in writing.
  • Manx: A Celtic language from the Isle of Man, revived after near extinction in 1974.
  • NKo: A standardized form of West African Manding languages with a unique alphabet created in 1949.
  • Punjabi (Shahmukhi): The most spoken language in Pakistan, written in Perso-Arabic script.
  • Tamazight (Amazigh): A Berber language from North Africa with mutually understandable dialects, supported in Latin and Tifinagh scripts.
  • Tok Pisin: An English-based creole and lingua franca of Papua New Guinea.

Choosing Language Varieties

Adding new languages involves careful consideration of regional varieties, dialects, and spelling standards. The company’s approach prioritizes the most commonly used varieties. For instance, the model for Romani produces text closest to Southern Vlax Romani, incorporating elements from Northern Vlax and Balkan Romani.

PaLM 2 plays a crucial role in efficiently learning related languages, such as those close to Hindi (Awadhi, Marwadi) and French creoles (Seychellois Creole, Mauritian Creole). As technology advances, and with ongoing collaboration with linguists and native speakers, we will continue to support more language varieties and spelling conventions.

Start translating today at translate.google.com or on the Google Translate app for Android and iOS.

Leave a Reply

Your email address will not be published. Required fields are marked *