Google added 110 new languages to its Translate. From Cantonese to Qʼeqchiʼ, these languages represent more than 614 million speakers, allowing translations to around 8% of the world’s population.
Some are languages spoken by more than 100 million people. Others are used by small indigenous communitiesand some have hardly any native speakers, but are being actively revitalized, Google explains.
These 110 new languages have been incorporated into Google Translate using the PaLM 2 advanced language model, which It is a type of artificial intelligence designed to effectively understand and generate natural language text.
These models are capable of learning complex language patterns from large amounts of data and can be applied in tasks such as machine translation.text generation, automatic responses, among other advanced uses.
Some of the languages recently supported in Google Translate are:
– Afar: Language spoken in Djibouti, Eritrea and Ethiopia. Of all the languages in this launch, Afar had the most voluntary community contributions, Google says.
– Cantonese: It has long been one of the most requested languages for Google. Because the language often overlaps with Mandarin in writing, it is difficult to find data and train models.
-Manx: Language of the Isle of Man, autonomous dependency of the British Crown in the Irish Sea, between England and Ireland. It almost became extinct with the death of its last native speaker in 1974. But thanks to an island-wide revival movement, there are now thousands of speakers.
-NKo: It is a standardized form of the Mandinka languages of West Africa that unifies many dialects into a common language. Its unique alphabet was invented in 1949 and today it has an active research community that develops resources for its learning.
– Punjabi (Shahmukhi): Most spoken language in Pakistan.
– Tamazight (Amazigh): Berber language spoken throughout North Africa. Although there are many dialects, the written form is generally understandable to everyone. It is structured in Latin and Tifinagh script, both compatible with Google Translate.
-Tok Pisin It is a fusion of English and the Papua New Guinea lingua franca.
To interact with Cantonese, Afar, Punjabi or the 107 new Google Translate languages, just follow these steps:
1. Enter Google Translate.
2. In the language bar at the top, select the language of interest.
3. Start interpreting texts, audios, documents and websites.
Linguistic varieties refer to the different ways in which a language manifests itself depending on the social, cultural, geographical or situational context in which it is used. A clear example of linguistic variety would be the Spanish spoken in different regions of Latin America and Spain.
Words like “computer” (Mexico) vs. “computer” (Spain), or the use of “vosotros” (Spain) instead of “ustedes” (Latin America), are examples of how linguistic varieties are manifested within the same language.
“Our approach has been to prioritize the most used varieties of each language,” says Google. “Our models produce text closer to southern Vlax Romani, a variety commonly used online. But it also mixes elements from others, such as Northern Vlax and Balkan Romani,” the company adds.
PaLM 2 played a crucial role in enabling Translate to improve its ability to more effectively learn closely related languages. This includes languages close to Hindi, such as Awadhi and Marwadi, as well as French creoles such as Seychellois Creole and Mauritian Creole.
“As technology advances and we continue to partner with expert linguists and native speakers, over time we will support even more linguistic varieties and spelling conventions,” Google concludes.
