Google Translate has added two more South African local languages to its library of languages it can translate. These languages are Tsonga and Sepedi which it can translate to and from English. It already featured other South African local languages such as Afrikaans, Sesotho, IsiXhosa and IsiZulu.
The two local languages were added together with other 24 local languages of other countries. This increased the number of languages that Google can translate around the world to 133 languages according to Google’s recent announcement.
The aim of Google to add these languages to their library of translations is to help break the language barrier and connect communities all over the world.
Using Zero-Shot Machine Translation, the newly additional languages are “a technical milestone.” This method allows for the creation of a translation engine for languages that lack translation dictionaries or significant amounts of translated content. Instead, it can learn a language by reading monolingual material from “under-resourced languages,” as the term is euphemistically referred as.
The language translations still need improvement
The translation of these local languages will not be perfect as yet as other European languages such as Spanish or German translation. But there is still room for improvement according Google Translate senior software engineer Isaac Caswell.
“While this technology is impressive, it isn’t perfect. And we’ll keep improving these models to deliver the same experience you’re used to with a Spanish or German translation, for example. If you want to dig into the technical details, check out our Google AI blog post and research paper,” said Isaac Caswell.
In a more detailed post on the artificial intelligence approach to adding the new languages, Caswell and colleagues were more blunt.
“We want to stress that the quality of translations produced by these models still lags far behind that of the higher-resource languages supported by Google Translate. These models are certainly a useful first tool for understanding content in under-resourced languages, but they will make mistakes and exhibit their own biases.”
Other African languages that Google Translate has got in its library
Google Translate has existing support for Afrikaans, Sesotho, isiXhosa, and isiZulu. According to businessinsider, Google translate also features a number of languages from the region and the continent, including Shona and Swahili. But many others are missing, and monolingual model-building currently seems the best bet for their inclusion.