Google Translate Now Adds 110 More Languages – Cantonese Included!

You certainly have used Google Translate at one point, but there might be rare cases where the language you’re looking for isn’t supported. In the “largest expansion ever” to the translation service, Google has added a whopping 110 new languages to its roster, and it includes Cantonese, a language/dialect that most Malaysian Chinese are familiar with.

Google Translate + LLMs

The mechanism of machine translations is very complex, and in some ways it isn’t too far fetched to say it is similar to LLMs (large language models) like ChatGPT. Of course, machine learning has gone a long way since its inception, and it’s a big help to Translate to keep the translations relatively accurate.

In fact, the latest additions are credited to Google’s PaLM 2 LLM, which helps Translate “more efficiently learn languages that are closely related to each other, including languages close to Hindi, like Awadhi and Marwadi, and French creoles like Seychellois Creole and Mauritian Creole.”

A few notable additions include Cantonese, which Google says is among the most requested languages but has been found difficult to implement and train due to its high similarity against standard Chinese in writing; others include Punjabi (in Shahmukhi script), N’Ko, Manx, Tok Pisin, Basa Bali, Minangkabau, and more.

As part of the company’s 1,000 Language Initiative announced in 2022, Google aims to built AI models that “will support the 1,000 most spoken languages in the world,” so this update is certainly showing promising progress in achieving the target.

Source: 9to5Google

Pokdepinion: I can definitely do with some translation help as someone who doesn’t speak Cantonese.