Allow "Announce" to use either AUX or RCA inputs as source

[tsheldon]

Example use case:

1. Person rings doorbell
2. Doorbell triggers automation
3. Automation triggers AmpliPi "Announce" from AUX input
4. Automation plays text to speech through a media device connected to AUX.
5. AmpliPi detects silence at end of announcement after "a small delay" in the same way it does at the end of the media file.

This would be a much more useful feature for PA rather than just being able to play an arbitrary (existing) media file.

[SteveMicroNova]

Hey tsheldon!

This sort of thing is somewhat supported at the moment with a few workarounds. One simple workaround is simply creating an audio file of whatever TTS you'd like to run such as "someone's at the door" and having the doorbell make the API request.
Another (potentially better) way is with our home assistant plugin, which allows for any valid HA TTS source to play directly to the announcement endpoint. This can save you from needing to stamp out an audio file and also gives you the power of home assistant's automation flows to have potentially more detailed information pending what you're interested in setting up.

As for specifically allowing aux to send stuff to announcements, I'll have to talk to the team about if that's something we're interested in doing, but I hope these workarounds serve you in the meantime either way.

[tsheldon]

Thanks for the quick reply!

One of the biggest reasons I'd love to see this implemented more at the hardware level is for those people leaning towards a more hardware based automation stack and less of a DIY / Home Assistant based approach. The particular use case I have in mind is something like Loxone that can generate arbitrary TTS through it's Audioserver product.

Even though I'm a "DIYer", if I'm integrating this level of product into a home, I'd like to have as few software based interconnects and possible- less to manually support and you can leave it integrated in the house if you ever move out as well.