GoKawiilEdgar Cervantes / Android Authority
TL;DR Google Deep Mind researchers have been developing tools to use AI to interpret sign language.
Now in Google’s latest Gboard update, we see the keyboard preparing to accept Sign-to-Text input.
Privacy protections would keep your video on device, and only share raw gesture data with Google’s cloud AI.
Google’s Gboard keyboard wants nothing more than to help you get the thoughts in your head turned into words on your screen, as easily as possible. We already have so many different ways to accomplish that, from good old-fashioned tapping away at letters, to the oh-so-slick “Glide” swipe input, to the modern convenience of voice typing. But now Google could be about to seriously level up its accessibility game, as Gboard gets ready to understand sign language.
You’re reading an Authority Insights story. Authority Insights brings you all the latest exclusive reports, app teardowns, leaks, and in-depth tech coverage from the Android Authority team that you won’t find anywhere else.
Being able to automatically convert sign language gestures into written text sounds almost fantastical, but with improvements in machine vision and AI processing, the stars are finally aligning to make this feat not just possible, but practical. Last year, Google Deep Mind teased its advanced SignGemma model for interpreting sign language input, and now we may finally be getting a look at one of its first major uses.
Checking out the changes present in version 17.8.3.939743344-beta-arm64-v8a of the Gboard app, we’re able to get a very early look at a new Sign-to-Text input option:
While we’re not yet able to get the tool actually working, this introductory pop-up gives us an early overview of what to expect.
AssembleDebug / Android Authority
... continue reading