{"id":3110,"date":"2020-07-13T18:44:07","date_gmt":"2020-07-13T18:44:07","guid":{"rendered":"https:\/\/translatorswithoutborders.org\/blog\/?p=3110"},"modified":"2024-08-22T14:28:32","modified_gmt":"2024-08-22T14:28:32","slug":"language-technology-initiative-levantine-arabic","status":"publish","type":"post","link":"https:\/\/translatorswithoutborders.org\/blog\/language-technology-initiative-levantine-arabic\/","title":{"rendered":"The latest from TWB\u2019s language technology initiative"},"content":{"rendered":"

Leaping over the language barrier with machine translation in Levantine Arabic<\/b><\/h2>\n

When a language you don\u2019t understand appears in your Facebook news feed, you can click a button and translate it. This kind of language technology offers a way of communicating not just with the millions of people who speak your language, but with millions of others who speak something else.\u00a0<\/span><\/p>\n

Or at least it almost does.\u00a0<\/span><\/p>\n

Like so many other online machine translation systems, it comes with a caveat: it is only available in major languages.\u00a0<\/span><\/p>\n

TWB is working to eliminate that rather significant caveat through our language technology initiative, Gamayun. We named it after a mythical birdwoman figure in Slavic folklore \u2014 she is a magical creature that imparts words of wisdom on the few who can understand her. We think she’s a perfect advocate for language technology to increase digital equality and improve two-way communication in marginalized languages.<\/span><\/p>\n

We have reached an important Gamayun milestone by leaping over the language barrier with a machine translation engine in Levantine Arabic. Here is how we got here, what we learned, and what is next.<\/span><\/p>\n

\"This
This is the Arabic word for \u2018security.\u2019<\/figcaption><\/figure>\n

What is behind developing a machine translation engine in Levantine Arabic?<\/b><\/h2>\n

In November 2019, we joined forces with a group of innovators and language engineers from PNGK and Prompsit to address WFP\u2019s Humanitarian Action Challenge. Our goal was to use machine translation to enhance the way aid organizations understand the needs and concerns of Syrian refugees, to improve food security programming.\u00a0<\/span><\/p>\n

So we developed a text-to-text machine translation (MT) engine for Levantine Arabic tailored to the specifics of refugees\u2019 experiences. To achieve this, we collaborated with Mercy Corps\u2019 Khabrona.Info team. The team runs a Facebook page for Syrian Arabic refugees to provide them with reliable information and answers, such as about accessing food and other support. We took content shared on the <\/span>Khabrona.Info Facebook page<\/span><\/a> and manually translated it into English to adapt the engine. The training data and a demonstration version of our MT are available on our <\/span>Gamayun portal<\/span><\/a>.<\/span><\/p>\n

How well does this machine translation engine perform?<\/b><\/h3>\n

To answer this question, we conducted an evaluation based on tests widely used by MT researchers. We found that our MT engine produced better translations for Levantine Arabic than one of the most used online machine translation systems.\u00a0<\/span><\/p>\n

We first asked experienced translators to rate the translations for both accuracy and fluency. We provided them with ten randomly selected source texts and translations generated by humans, Google\u2019s MT, and our MT. All translations were fairly good, with scores ranging from zero for no errors to three for critical errors. Our MT engine performed slightly better than Google\u2019s MT because it was adapted to the specifics of Levantine Arabic and its online colloquialisms about food security and other topics relevant to refugees\u2019 experiences. The human translations performed slightly better than our MT, but were not perfect.\u00a0<\/span><\/p>\n