I’m sure this has been covered, but I spent several hours and couldn’t find a solution.
I set up a Wyoming satellite on a Raspberry Pi 4 with a usb mic and a speaker plugged in to the native speaker output. I have whisper and piper set up, as well as Openwakeword.
The wake word works fine. And debug shows my speech to text is working fine. The debug even shows that TTS output has been created. Verified by cached wav files in the TTS folder on my HA system. But no sound is ever heard.
I can go to the pi and manually record and play a wav file, and I can use automations that use TTS and they both work fine. It is just the assistant that is not able to broadcast TTS. Ideally I would like to use the speaker connected to my PC rather than the speaker on my Pi as it sucks. But at this point I’ll take sound from anything at all.
Please understand that I am still VERY new to HA so if you provide a script, please tell me where to put it. I looked in Studio Code Server but the only things I could find that were related to the assistant were the TTS folder and a couple log folders.
Any assistance would be greatly appreciated.
Regards,
Tony
HA is running on a Nuc with all updates installed as of 8-23-2024