Google has rolled out its latest Pixel feature drop, giving Pixel 6 and newer users home screen widgets for quick calls or texts to VIP contacts, plus new scam detection warnings built into the Gboard keyboard for chat apps. The update also brings Harry Potter-themed customization packs and two free months of Audible, tied to the release of a new full-cast audiobook series, alongside gesture improvements for the Pixel Watch line.
cnet.com
· 2026-09-15
Google rolled out its latest Pixel Feature Drop, adding expanded home-screen widgets and notification badges for quickly reaching VIP contacts on Pixel 6 and newer. The update also broadens Scam Detection across chat apps and Gboard in ten countries, brings a Wear OS 7 update with offline Gemini and improved GPS to Pixel Watch 5, and introduces Harry Potter-themed customization packs with free Audible access.
engadget.com
· 2026-09-15
A Reddit demo shows the Pixel 11's American Sign Language translation feature converting continuous signing into English text within Gboard, confirming it works as advertised beyond Google's official showcase. The tool, powered by Google's SL2T model, analyzes hand, arm, torso, head and facial movements to interpret meaning rather than requiring manual input.
androidauthority.com
· 2026-09-07
Google has quietly updated its developer documentation to reveal that just seven Android phones currently support Gemini Nano 4: the Pixel 11, Pixel 11 Pro, Pixel 11 Pro XL, Pixel 11 Pro Fold, and Samsung's Galaxy Z Flip 8, Z Fold 8, and Z Fold 8 Ultra. These are the only devices officially equipped to run Google's latest on-device Gemini Intelligence features.
androidauthority.com
· 2026-08-28
Google has launched Gemini 3.5 Transcribe, a new voice-to-text AI model that automatically strips out filler words and stumbles to produce cleaner transcriptions. Already running Gboard's Rambler feature on the Pixel 11, the model is set to expand across Google's products. Google says it is roughly 70 percent faster than its predecessor, Chirp 3, and slightly more accurate, with a live-speech error rate of 5.5 percent versus 7.32 percent.
arstechnica.com
· 2026-08-26
GoKawiil