Facebook Introduces AI-based Machine Learning Software Than can Translate Over 100 Languages

Facebook, the social media giant has rolled out software-based on machine learning. The popular social networking platform stated that it was the first firm to translate from any of 100 languages sans depending on English. The new artificial intelligence software unveiled by Facebook is open-source software. The company has created this software to assist the huge social network to deliver better content in 160 languages to more than 2 billion users globally.

Angela Fan, the research assistant at Facebook said in a blog post that, ‘This milestone is a culmination of years of Facebook AI’s foundational work in machine translation.’ Angela further stated that the new model is extremely precise compared to the other systems. The reason is that the new model doesn’t depend on English as an intermediary translation step. 

Facebook Introduces AI-based Machine Learning Software
Facebook Introduces AI-based Machine Learning Software

In the blog post, Fan wrote, ‘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.’ The new machine learning model directly trains on Chinese to French information to sustain its meaning. It surpasses English-centric systems by 10 points on the prevalently accessed BLEU metric for assessing machine translations, said Fan.

On its news feed, Facebook stated that it has been already handling an average of 20 billion translations daily. The company wishes the new system will render better results compared to the existing systems. On the blogpost, Fan said, ‘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.’

Stay tuned to this website to get the latest tech updates, telecom news, latest technological advancements, and more.

Leave a Reply

Your email address will not be published. Required fields are marked *