Oh no - I changed email providers yesterday! Let me look into that.
I am pushing some bug fixes around wake word detection and multithreading.
Also adding support for Sendspin to make these capable multiroom audio targets. Really excited for that project and the native support for it in Music Assistant!
Thanks @ecohash-co, that’s awesome.
Would love to join the beta, but I’m currently traveling and don’t have access to my device…
Sendspin support would be really amazing - can’t wait to be able to add all my players into one sync group.
From the demo and the description it seems a very capable solution!
I know that having spent many hours of work on it the closed source seems a good choice and I understand the motives, but….
The one reason I wouldn’t use FKB or other closed source alternatives is that I don’t want any sw not open running 24/7 in devices with mic and cameras in my home.
If this project will be open in the future I ‘d surely take a look at it!
It is added! I freaking love it. It is so much easier than snapcast and works great for multi-device sync.
I am still trying to get onscreen player controls working in DashVoice and then will add UI elements to manage whole house audio.
1 Like
I love and share that mindset – you can absolutely lock down a tablet running DashVoice (i.e. cut it off from the internet) and it will still work great.
It would need Internet to download the local voice model during onboarding and it would still need access to your local homeassistant server, but other than that you could cut it off from the internet at your firewall.
You could occasionally reconnect it to pull new updates from the Play Store, but local only is a real thing here.
3 Likes
I am interested in testing this.
nice – email [email protected] and we’ll get you added to the testers list so you get enrolled and can download from the Play Store.
1 Like
Being closed source we simply can’t know what the app does - will (private) recordings be uploaded to the cloud like it is the case with amazon Alexa? Yes/No? Does not really matter as we can’t read/build the sources because they are closed 
Seems like some very capable open source alternative already exists 
Tried the app out on a hacked Echo Show 8. I dont run piper (i use kokoro with a wyoming proxy) so TTS didnt work. Couldnt get wale word working either unfortunately. The idea is promising but seems a bit too buggy (and closed source) at the moment. Keep up the hard work!
Logcat? Thanks!