Can I run both and slowly transition to the new one so that I don't have to get everything right real quick so the significant other doesn't squawk?
I transitioned from beta to stable. Not much to setup, music providers and player providers. I can't remember the task being major.
but the new container is still music assistant, just another version. Or is beta channel using diffeerent app name?
Different app name. You can have up to four containers on HA simultaneously (just not running at the same time) - Stable, Beta, Nightly, Dev
Hi all, forgive me if this is not the right place to ask (though it seems to be based on my searching).
I'm hoping someone may be able to provide me a bit of direction. I am running MA as an integration in HA. This was set up with my original HA owner/admin user, logging into MA with the internal HA ingress.
I have been playing around with remote access to HA through Netbird and have gotten everything working nicely. Within HA I have added the OIDC/OAuth2 integration from HACS and am able to access HA locally and remotely using Authentik (with the newly auto-created user in HA) as my IdP.
I am able to access MA when logged in as the original owner user though HA ingress but have found that MA cannot connect/authenticate when logged in through Authentik.
I guess my primary question needs to be, is MA even able to connect to HA via an external IdP like Authentik? Would I need to recreate my MA server while logged in with my Authentik user if so? Or would I need to set up or add a MA-specific redirect URL for Authenik?
At this point, my gut is telling me to ditch Authentik with HA all together and just rely on the built-in HA authentication for local and remote access.
Apologies for any vagueness here. I'm currently trying to wrap my head around where to start looking to solve the problem.
Hi, is support for playing webm files planned for the future?
best regards
Jarny
Just do this.
Fair enough! Thank you both!
Is it possible to create permament groups yet in MA?
I've been using media_player.join and media_player.unjoin to create dymanic groups if I want a TTS announcement to play on multiple speakers. Previously, if I left media players in a group and tried to play music on one player in the group, it played on the whole group and I would frantically have to run a media_player.unjoin action to stop it playing on the group.
I wondered if that was no longer the case and I could now create permanent groups that also allow playback on one player in that group? Kinda like in the Amazon Echo ecosystem where I could group my Echos by floor or whole house etc and leave those as permanent groups while still being able to play content on a single Echo.
At the moment I can ask Assist to "Play track by artist in the kitchen" and it will do so. I want to be able to say "Play track by artist everywhere, or downstairs and have it do that too.
Hope I'm making sense.
The short answer is yes. Frequently Asked Questions - Groups - Music Assistant
To play an announcement to an individual player the group must be powered off or the individual player must be removed from the group.
1 Like
Thanks, that's good to know.
On my intial skim through of that section I found this:
NOTE
If an attempt is made to power on a group and a member of that group is already a member of another powered on group then an error will be shown.
So all of my media players are HA VPEs. My TTS announcement system's speech engine is setup to play any announcement on the VPE in the room that I am in ie it finds me via a complex series of scripts that output the entity_id of the VPE in the room I am in.
If I create a permanent group and that group is playing music when my speech engine tries to play an announcement to one of the VPEs in that group, will the announcement fail?
Secondary question, would I therefore need to develop my speech engine such that it detects if a VPE it wants to play an announcement on is being used by a group and either stops the group play so it can use that VPE or sends the announcement to another VPE?
I havent played with announcements for a while so the easy answer is just to try it and see what happens. Looking at the docs it seems like it should work Announcements - Music Assistant because sendspin has native announcement support
1 Like
Thanks. The docs are really useful, but the one thing that stands out for me is:
Announcements are either Text-to-Speech or audio files that are sent (from Home Assistant) to the MA players via the tts.xxxxxx or music_assistant.play_announcement HA actions.
Overview
If music is being played by MA then this will be temporarily paused and restored after the announcement.
I use chime_tts.say for all my announcements on HA VPEs and yes, the content being played by MA stops when an announcement is played, but it does not resume after the announcement concludes. So I guess Chime TTS is not supported.
That in itself is not an issue, because through the power of Home Assistant, anything is possible and I've created a way to mimic the Music Assistant play resumption through my speech engine script.
If there isn't a feature request for the inclusion of chime_tts.say I'm happy to create one, but I don't want to create extra work for you if I can solve this locally.
Music Assistant is working really well for us and has been instrumental in allowing me to ditch the Amazon Echo devices, so I appreciate all the work that goes into this.
MA doesn't support custom components. However, it does support ducking of volume on devices that natively support that such as Sendspin, and Sonos. You can also choose not to send the MA chime.
Music Assistant recently crashed and wouldn’t launch. Are you having this problem too?
I have to press the start button to get it to start.
No one else is reporting this. For support ask on discord or GitHub.
I would restore from backup (just MA)
I’ve just noticed that all my custom playlists are missing in MA. They aren’t being autodiscovered, and if I manually import one to test, the playlist file imports fine, but all the listed tracks are showing as ‘no playable items found’, even though I can browse and find the track manually and it plays just fine…
Any ideas ?
P.S. It all used to work fine, and I haven’t changed anything about my setup.
I’m using a remote SMB share which holds both the playlist files and the music tracks. This hasn’t changed.
Just incase its relevant, the playlist files contain paths relative to the root of the SMB share, not full paths, which has always worked in the past.
Seems to be a known issue:-
If I browser and play the playlist file, it finds all the tracks and plays them fine.
No that issue is for Jellyfin. Are you running the beta?
Not running Beta, but my symptoms are exactly the same.
Autodiscovery does not work for m3u playlists on my SMB remote share.
Importing playlists ‘works’ but no media is playable, it all says ‘no playable items found’
If I browse to the playlist it plays fine.
If I ‘add to library’ then the playlist is also fine.