I’m trying to build multiple Alexa show replacements with Pi’s. I’ll try to explain the process below.
Satellite HA touch screen > use Rhasspy wake word like Jarvis > send STT to Ada > Ada performs the request > Ada’s TTS response goes to AWS Polly > Polly sends TTS audio feed back to originating HA terminal
Not sure if that is the exact process flow that is most efficient to do what I’m looking for. I’m a first time user and I’ve already spent 100 hours trying to learn how this all works. I built a lab HA with a NUC running HASS image with Rhasspy and integrated the two. I was able to successfully build the slots and sentences for voice commands to wake up and perform the commands I told it to. I’m having trouble figuring out how to spread it out over multiple devices though on top of using Polly.
I understand this is a massive venture but if I can at least get an idea of what to expect or just help to make a portion of this work, I can piece it together better. Anything would be much appreciated!