Hi everyone!
The new Agent IA brick unlocks a lot of things for Gladys, and in particular the « Voice Command » part becomes very interesting 
I made a small proof of concept that I wanted to share with you:
If you want to test it, I provide a Docker image:
gladysassistant/gladys:gladys-stt
As for Speech To Text, I use an Open-Weight model (whisper-large-v3) hosted on Scaleway in France!
I think this feature could be practical if you have a wall-mounted tablet and you want to send a quick command