Google Home gets smarter voice recognition update
▼ Summary
– Google Home’s facial recognition will now use non-biometric signals like body size and clothing color to identify people even when their faces are not clearly visible, starting June 23rd.
– The Familiar Faces library will automatically update with recent images of household members to reduce inaccurate notifications from outdated photos.
– The update aims to reduce the likelihood of smart home cameras misidentifying people based solely on their back or side views.
A new update for Google Home aims to cut down on those frustrating moments when your smart home cameras fail to recognize you simply because you’re not looking directly at them. Starting June 23rd, Google is expanding its facial recognition feature to identify people in your Familiar Faces library even when their faces aren’t clearly visible. The system will now rely on “additional non-biometric signals” such as body size, clothing color, and other contextual cues to keep tagging the right person.
The Familiar Faces library will also become more dynamic, automatically updating with the most recent images of everyone in your household. This means you should receive fewer inaccurate alerts caused by outdated reference photos. By refreshing the database on its own, the system adapts to changes in appearance over time, reducing the need for manual updates.
This enhancement is part of Google’s broader effort to make its smart home ecosystem more reliable and user-friendly. For anyone who has ever been locked out of a routine or received a false notification because a family member was walking away from the camera, this update promises a smarter, more consistent experience.
(Source: The Verge)
