fbpx
Qatar v. Ecuador to kick off FIFA World Cup 2022™ on 20 NovemberRead more Webb Fontaine Announces Launch of Niger National Single Window (NNSW) to Bolster TradeRead more Ethiopia: Loan from United Nations Fund Allows Food and Agriculture Organization (FAO) to Scale Up Fertilizers for Farmers in TigrayRead more How Choosing the Right Printer Helps Small Businesses and Content Creators to Save Time, Maximise Productivity and Achieve GrowthRead more The United States Contributes USD $223 Million to Help World Food Programme (WFP) Save Lives and Stave Off Severe Hunger in South SudanRead more Eritrea: World Breastfeeding WeekRead more Eritrean community festival in Scandinavian countriesRead more IOM: Uptick in Migrants Heading Home as World Rebounds from COVID-19Read more Network International & Infobip to offer WhatsApp for Business Banking Services to Financial Institution Clients across AfricaRead more Ambassador Jacobson Visits Gondar in the Amhara Region to Show Continued U.S. Support for the Humanitarian and Development Needs of EthiopiansRead more

Facebook unveils machine learning translator for 100 languages

Print Friendly and PDF

Oct 20, 2020 - 06:23 AM

WASHINGTON — Facebook on Monday unveiled software based on machine learning which the company said was the first to be able to translate from any of 100 languages without relying on English.

The open-source artificial intelligence software was created to help the massive social network deliver content better in 160 languages to its more than two billion users around the world.

“This milestone is a culmination of years of Facebook AI’s foundational work in machine translation,” research assistant Angela Fan said in a blog post.

Fan said the new model is more accurate than other systems because it does not rely on English as an intermediary translation step.

“When translating, say, Chinese to French, most English-centric multilingual models train on Chinese to English and English to French, because English training data is the most widely available,” Fan wrote.

“Our model directly trains on Chinese to French data to better preserve meaning. It outperforms English-centric systems by 10 points on the widely used BLEU metric for evaluating machine translations.”

Facebook said it already handles an average of 20 billion translations every day on its news feed, and that it hopes the new system will deliver better results.

“Breaking language barriers through machine translation is one of the most important ways to bring people together, provide authoritative information on Covid-19, and keep them safe from harmful content,” Fan said.

ZONNTECH.COM uses both Facebook and Disqus comment systems to make it easier for you to contribute. We encourage all readers to share their views on our articles and blog posts. All comments should be relevant to the topic. By posting, you agree to our Privacy Policy. We are committed to maintaining a lively but civil forum for discussion, so we ask you to avoid personal attacks, name-calling, foul language or other inappropriate behavior. Please keep your comments relevant and respectful. By leaving the ‘Post to Facebook’ box selected – when using Facebook comment system – your comment will be published to your Facebook profile in addition to the space below. If you encounter a comment that is abusive, click the “X” in the upper right corner of the Facebook comment box to report spam or abuse. You can also email us.