cosmic zenith
I found a GPU accelerated kokoro at image: ghcr.io/remsky/kokoro-fastapi-gpu:latest. I know very little other then kokoro sounds very nice.
Anyone know of or using a docker image that is GPU accelerated but has wyoming exposed? I need HA to see it so I can use it with the new voice preview hardware that just showed up today 🙂
deft marlin
I haven't had a chance to test it yet but this can work as middleman: https://github.com/roryeckel/wyoming_openai
There's even an example here: https://github.com/roryeckel/wyoming_openai/blob/main/docker-compose.fastapi-kokoro.yml
tiny flower
Since Kokoro probably won't use all of your VRAM, this will also include Whisper for STT: https://github.com/speaches-ai/speaches - use the wyoming_openai repo that Impact mentioned to make it work with Home Assistant.
deft marlin
Example compose file here: https://github.com/roryeckel/wyoming_openai/blob/main/docker-compose.speaches.yml
cosmic zenith
Awwesome!
I need to dodge faster-whisper already using some ports but i think i can just use docker port mapping
cosmic zenith
I have it all setup, but wyoming_openai is complaining about needing an API key. :/
openai.AuthenticationError: Error code: 401 - {'error': {'message': "You didn't provide an API key. You need to provide your API key in an Authorization header using Bearer auth (i.e. Authorization: Bearer YOUR_KEY), or as the password field (with blank username) if you're accessing the API from your browser and are prompted for a username and password. You can obtain an API key from https://platform.openai.com/account/api-keys.", 'type': 'invalid_request_error', 'param': None, 'code': None}}
Got it sorted!