Google Home now recognizes you by body size, clothing color
What's the story
Google is enhancing its facial recognition capabilities for the Home app, making it less likely for smart home cameras to misidentify users. The update, which rolled out yesterday, expands the Familiar Faces library feature. It will now use "additional non-biometric signals (body size, clothing color, etc.)" to identify tagged individuals even when their faces are not clearly visible.
Auto-update
Familiar Faces library gets auto-update feature
Along with the enhanced recognition, Google is also improving the Familiar Faces library by making it auto-update with the latest images of all household members. This way, users will get fewer inaccurate notifications from outdated examples. The move is part of Google's effort to improve user experience and reduce errors in its smart home system.
Sound recognition
AI-generated video event descriptions improved
Google is also improving its AI-generated video event descriptions. The feature will now be able to identify specific sounds such as dogs barking, alarms, or footsteps and include them in the notes. This works even if the audio was from something off-camera. The move is aimed at making the smart home system more accurate and reliable for users.
App enhancements
New 'system health alerts' in Google Home app
The latest update to the Google Home app, version 4.20, brings new "System Health alerts." These alerts are triggered when a Nest thermostat detects an issue with the HVAC system. The feature appears to be a Gemini-connected upgrade of the existing Google Nest System Health Monitoring. The update also brings improved support for Matter switches, further enhancing user experience and functionality of the app.