Hello,
Over the last couple of years I’ve kitted my whole house with smart speakers from the Google Nest range. I also have GoogleTVs. All of those units are fully integrated into homeassistant so I can control them etc.
A few months ago I started experimenting with the OpenAI integration to the local HA-Voice assistant and I’ve simply loved the integration and the responses. It seems that it simply understands what I am asking and giving back responses that are relevant. However this is only by writing to it. Now I want to use my smartspeakers.
I’ve searched for ages now on how to reuse my Google nest speakers (mainly microphone and speaker) but simply route the “request” to HA-VoiceAssistant so that I can leverage my OpenAI configuration. It would also make it so much easier to only have to expose/configure my devices in 1 place instead of several. Does anyone have any clue how I can achieve this?
If I would have bought all the smart speakers today I would have just bought Home Assistant Voice, but I invested in tons of speakers even before Homeassistant released their own hardware.
TLDR; I want to use my Google speakers but bypass their “AI” and instead route all the requests/responses to/from Homeassistant Assistant so that I can leverage my configuration of OpenAI. I have no issues with latency or that all my data will be sent to OpenAI.