Replies: 1 comment
-
Some issues as I see them:
Anything supporting multi-user (either via audio or visually) is inherently going to drastically reduce the user experience of interacting with Willow and the platforms it supports. Not to say that we'll never do it via some means but generally speaking I also like to say that we focus on real-world utility, usability, practicality - not cool demos/party tricks. The Willow Inference Server could potentially be extended to detect multiple users via voice eventually but even that would have to be REALLY solid for us to include it. That said, of course Willow is open source and I would actually be entertained by people adding functionality such as this, but in many/most cases they likely wouldn't make it in tree. |
Beta Was this translation helpful? Give feedback.
-
Hi,
There's mentions of multi users, just an idea on one of the ways this may work is face recognition. There was a post on reddit recently about someone who managed to get that working using a cheap (relatively speaking) mmwave sensor on an ESP32, not sure how heavy that is and if it could run at the same time as Willow but that could be one use for the GPIO pins.
The thread in question : https://www.reddit.com/r/homeassistant/comments/13mru0s/facerec_beta_simple_fast_local_and_secure_face/
Feel free to close this if it doesn't align with the vision, just a thought :)
Beta Was this translation helpful? Give feedback.
All reactions