Saturday, July 20, 2024

Indigenous languages including Maya, Zapotec and Nahuatl added to Google Translate

Google has added 110 new languages — including Zapotec, Nahuatl, Maya Yucateco and Qʼeqchiʼ, also a Maya language — to its free Google Translate service.

Mexico is one of the most linguistically diverse countries in the world, with 68 Indigenous languages spoken in the country by 7.5 million people. However, almost 300 Indigenous languages ​​in Mexico have already disappeared.

Machine translation is valuable in the efforts to save minority languages, which could help preserve Mexico’s Indigenous languages.

Indigenous Mexican languages now supported by Google Translate

Zapotec is an extensive language family originating in southern Mexico. There are over 50 different Zapotec languages (which belong to the larger Otomanguean language family). Mexico’s national statistics agency, INEGI, reports that a total of approximately 425,000 people speak Zapotec, primarily in the state of Oaxaca.

Nahuatl (also known as Aztec, or Mexicano) is a group of languages of the Uto-Aztecan language family. About 1.7 million Nahua people speak Nahuatl, primarily in Central Mexico.

A screenshot showing a Nahuatl translation on Google Translate
Nahuatl, Maya, Zapotec and other Indigenous languages spoken in Mexico are now available on Google Translate. (Google)

Qʼeqchiʼ (Kekchi in many English-language contexts, such as in Belize) are a Maya people of Mexico, Guatemala and Belize. Their language is Qʼeqchiʼ.

Maya Yucateco, often simply called Maya, is spoken in the Yucatán Peninsula and parts of northern Belize. Linguists added “Yucateco” to the name in order to clearly distinguish it from all other Maya languages.

Google Translate’s latest expansion is its largest yet

In June, Google announced its plan to add 110 new languages to Google Translate, its largest expansion yet.

The expansion is part of a Google initiative to build AI models to support the 1,000 most spoken languages around the world.

“We’re always applying the latest technologies so more people can access this tool … From Cantonese to Qʼeqchiʼ, these new languages represent more than 614 million speakers, opening up translations for around 8% of the world’s population,” the tech giant said in a statement.

These languages are in different stages of usage. Some have 100 million speakers, Google noted, and some have no active speakers.

In determining which languages to provide support for in Google Translate, Google said it considers regional varieties, dialects and different spelling standards, the news site Tech Crunch reported.

A large language model called PaLM 2 powers the new expansion. PaLM 2 was pre-trained on multilingual sets of texts ranging in scope from human to programming languages, making it much more advanced than its predecessor PaLM, which used mostly English-only datasets.

Today, Google Translate supports 243 languages at various levels.

With reports from El País, Tech Crunch and The Register

Have something to say? Paid Subscribers get all access to make & read comments.
El Chavo del 8 and La Chilindrina

The story of ‘Little Shakespeare’ and a generation of Mexican comedy

Who was the man behind Mexican TV's most iconic characters and what was the complicated legacy he left behind?
A rapt audience in the dark watches a film at last year's Guanajuato International Film Festival (GIFF)

Guanajuato International Film Festival opens its 27th edition

The festival will show nearly 200 films, including 16 that are making their world premiere.
201st Airborne, Mexican airforce WWII

The Aztec Eagles: Mexico and the Second World War

Often overlooked, Mexico's brave airmen saw combat alongside the United States in the war over the Pacific, winning the respect of their American colleagues.