Meta develops AI system that translates 200 languages ​​in real time

Deepak Gupta July 7, 2022
Updated 2022/07/07 at 8:11 PM

Meta, the company that owns Facebook, has long expanded its horizons as a “social network”. The creation of a metaverse, for example, wants to put people to live “in society” a parallel world through virtual reality. The language barrier will also have to be overcome so that real connections can be created and an efficient translator will be essential.

That’s why, through a new model of Artificial Intelligence (AI), Zuckerberg’s company is developing a never-before-seen translation system.

Currently, translation tools already achieve an incredible reach and with a smartphone, even without an internet connection, it is already possible to establish communication in many places, but the truth is that the limitations remain great.

AI system that does real-time translation of 200 languages

The AI ​​translation system being developed by Meta, called NLLB-200 (from "No Language Left Behind"), aims to "no language is forgotten", in order to reach as many people as possible, without there being any a demarcated language barrier.

To help people communicate better today and be part of tomorrow's metaverse, AI teams created No Language Left Behind, an effort to develop high-quality machine translation resources for most of the world's languages.

The NLLB-200, according to Meta, already translates 200 different languages ​​with far more accurate results than previous technology could achieve. When comparing the quality of translations to previous AI surveys, the NLLB-200 scored a higher average of 44%. For some African and Indian languages, the NLLB-200 translations were over 70% more accurate.

To evaluate and improve the NLLB-200, the FLOWERS-200, a dataset that allows researchers to assess the performance of this AI model in 40,000 different language directions. FLORES-200 allows measuring the performance of the NLLB-200 translator in each language to confirm that the translations are of high quality.

Both systems are open source, so that more people can contribute to their development and improvement of the translations. In addition, there is a $200,000 fund to be invested in non-profit organizations and researchers working on initiatives focused on sustainability, food security, gender-based violence, among others, that operate in African countries and that use the NLLB-200 to communicate in two or more local languages.

This work will make it possible to further improve the reliability of the translations.

The AI ​​system will also support the more than 25 billion translations made daily on your most popular social networks.

!function(f,b,e,v,n,t,s){if(f.fbq)return;n=f.fbq=function(){n.callMethod?
n.callMethod.apply(n,arguments):n.queue.push(arguments)};if(!f._fbq)f._fbq=n;
n.push=n;n.loaded=!0;n.version='2.0';n.queue=[];t=b.createElement(e);t.async=!0;
t.src=v;s=b.getElementsByTagName(e)[0];s.parentNode.insertBefore(t,s)}(window,
document,'script','https://connect.facebook.net/en_US/fbevents.js');
fbq('init', '1664527397186427'); // Insert your pixel ID here.
fbq('track', 'PageView');
(function(d, s, id) {
var js, fjs = d.getElementsByTagName(s)[0];
if (d.getElementById(id)) return;
js = d.createElement(s); js.id = id;
js.src = "//connect.facebook.net/pt_PT/sdk.js#xfbml=1&version=v2.3&appId=122308327859118";
fjs.parentNode.insertBefore(js, fjs);
}(document, 'script', 'facebook-jssdk'));

Share this Article
Leave a comment

Leave a Reply

Your email address will not be published.