synesthesiam
commented
4 years ago Speaker recognition would be a welcome addition to Rhasspy, but is not currently implemented by any of the speech to text systems. Kaldi's x-vectors are probably our best shot. Not something I've had time for, unfortunately :/
Hi,
I'm trying to automate my house and want to use rhasspy as a voice assistant.
I want to have a mic in every room and therefore I need the option to use multiple microphones as inputs. However, I did not find this option in the settings or elsewhere on the internet.
Furthermore, I could not find any information how I could differ human voices, i.e. the system recognizes who I is speaking and adjusts the output respectively.
TY for any answers. Have a nice day!