Google Translate gets improved offline translations with the help of AI

Adjust Comment Print

The example Google gave (below) shows how the AI translation does make a difference, although the result still isn't quite as good as the online version.

A couple of years ago, Google added the Neural Machine Translation (NMT) technology to its Translation app, a step that was aimed at increasing the accuracy of translations. Google says the translations are better because the AI can take an entire sentence into context rather than translate piece-by-piece, which would end up with more literal and likely not as natural translations. According to Google Translate Product Manager Julie Cattiau, the technology takes a nuanced approach by leveraging artificial intelligence to take a whole sentence into context instead of translating it piece-by-piece, making the translations better.

Google Translate's offline feature can be used if you're traveling internationally, don't have access to the internet or don't want to use your data.

Now, how often will you need to use Google Translate without an internet connection? Google even provided a comparison of how the quality of translations should soon improve for offline users. That makes translations of paragraphs and articles a lot smoother, the company said in a blog post. Language sets are only 35-45MB in size and also run on lower-powered devices. Hopefully, the results will generally be close enough to help you get by. The end result should be offline translations that are more accurate and almost on par with the quality users expect when an internet connection is available.

The NMT update will be rolled out in 59 languages, including several European and Indian languages, over the next few days.