Microsoft Translator gets offline AI-powered translations

Microsoft is making its Translator apps even more useful when you don't have an internet connection, thanks to a little AI help. The tech giant announced today that the Microsoft Translator app on Android, iOS, and Amazon Fire devices now feature offline AI-powered translations, boosting their accuracy through the use of new offline language packs.

Microsoft first rolled out online AI-powered neural machine translation (NMT) in 2016 before adding the same capability to Android phones with dedicated AI chips in 2017. Now, Microsoft says the team behind its Translator apps was able to tweak its algorithms to work directly on any device's hardware.

"Building on this initial work, the Translator team was able to further optimize these algorithms, allowing them to run directly on any modern device's CPU without the need for a dedicated AI chip," Microsoft says. "The new Translator apps now bring NMT to the edge of the cloud for all Android, iOS*, and Amazon Fire devices." Windows devices are also expected to get support for AI offline translations soon.

The updated language packs are available for Arabic, Chinese Simplified, French, German, Italian, Japanese, Korean, Portuguese, Russian, Spanish, and Thai. More languages are planned to support offline AI-powered translations soon, but Microsoft is already promising the updated packs are up to 50 percent smaller while offering translations that are up to 23 percent better.

Aside from the new packs, Microsoft revealed that Android developers can now take advantage of these offline capabilities in their own apps. Third-party apps can take advantage of a new local feature that allows them to tap into Translator, relying on its offline capabilities to serve up translations without an internet connection, or retrieving the translation from the cloud when connected. The feature is currently in preview and is expected to launch in general availability in the coming months. If you're a developer, you can learn more at GitHub.

See at Google Play See at the App Store