Long after your add-on was released, Google made it easier to control media players.
Yesterday I experimented with the Google / HA (local install not Nabu Casa or Docker) out-of-the-box and I have basic pause/resume & volume working.
I know that your add-on adds a lot more.
My question … with the newish capabilities from Google Assistant for controlling media players would it be feasible to use their capability without having to go through Dialogue Flow?
If the HA built-in capability include treating LMS favourites like Input Selectors then I think I would not need anything more and am tempted to see if I can add that into the standard HA.
Then I could say something like
set input to preset 4 on bedroom radio
to play BBC Radio 4 on a bedroom Squeezebox.
Hello,
Anyone can say me if it’s possible to add more than 10 squeezebox player to this addon ? I’ve 15 squeezebox player. If it’s possible, someone can explain me how i can do that?
Thanks
Certainly is possible. Modify the lmscontrols.yaml to include the additional players, names, pull down selections and make the minor script modification following the same format for the first 10 players for the sync function. The yaml file is pretty readable walk through it slowly and I’m sure you’ll manage.
Ynot.
Not sure how exactly you’re doing the local control from google so kinda hard to comment. Perhaps provide a link to the technique you’re using and I’ll take a look. Ynot.
I was thinking that the media selector could be used to map to LMS favourites / presets and would work without dialogue flow since it is built in to GA nowadays.
Also, in theory, the new local control capability from GA (where JavaScript is loaded into the local GA device would also be interesting but I think that is/was being worked on in HA itself.
Hi Paul, Using Google Actions is basically what happens when you use Dialogflow. Dialogflow authors the intent and allows you some addtional flexibility to determine what exactly the users request means and multiple parameters (ie: song, player name, random status all in one request). Once that code is complete, its sent to Google actions for implementation. As for local control using Java I have never investigated that.
I’ve been trying for the last few days to make my install of LMS Controls work with my HA installation but with no success. The “Logitech Squeezebox” integration that’s natively supported by HA works perfectly fine and I can control my players with it.
On the other hand, I cannot figure out what is wrong with my LMS Controls. I’ve activated the homeassistant.components.media_player.squeezebox logs but I don’t get what the issue is from looking at the logs.
I’ve read the Installation Instructions.md a few times, going over every step and double-checking my config files. The troubleshooting.md file give the same instruction to test the basic installation using the GUI front-end but as mentioned, that fails.
After finding the native Sonos integration super underwhelming, about three or four months ago I started messing with LMS in my sandbox, and started using it for real about six weeks ago. I’m not sure exactly when I came across this guide, but since voice assistant media queries have been the most desirable and elusive Home Assistant functionality to me, as soon as I found it I added it to my config.
The player controls, sync group controls and media queries all worked great from Lovelace, although as @PaulWebster mentioned, the player and sync group controls are also functional through the native LMS and Google Assistant integrations. While the media queries were great to have in Lovelace, the main idea was to get them working with Google Assistant, which is where the trouble started.
After following the steps to import the dialogs into a new agent DialogFlow, tests in the DF console were successful but the same tests failed in the Assistant console and on my actual Assistant devices. I couldn’t produce any informative logs but triple-checked every possible setting in comparison to a working agent I have for a basic temperature reading, and couldn’t find any difference. I know this issue could well be unrelated to this specific config but I was wondering if there was anything I should try?
Hi Its been a while since I played with this setup. I do believe when you publish your control, its takes a fairly long time to make it accessible to your google devices (hours). The Assistant console where you are logged in as the developper should work immediately however if my memory serves even without publishing it. Will try to give this a shot sometime this week and let you know if I run into anything.
Not sure if you’ve solved this issue but I recently did do a full re-install of HA, imported project, new dialog flow, google actions, google home, etc. and everything seems to be functioning well. I can test the agent or call it from google home and my google mini’s.
Ah, thanks for confirming! Do you have any other HA integrations using agent functionality, like the more basic Google Assistant integration, Google Calendar etc? It seemed to be a matter of fixing one thing breaks another, but I’ve kinda stalled on DialogFlow stuff for the moment as I removed agents I thought were obsolete but were used for Calendar and Fit, and I lack the motivation to deal with my own stupidity right now
I played a bit with Google assistant directly (rather than dialogflow) but was having trouble parsing the json replies properly. Dialogflow is just much easier (at least for me) at this time. If I discover some spare time, I will give it another try with a simpler json payload.
@ynot, is it able to use it without dialogflow? I am fairly new to squeezebou/lms and would like to use rhasspy for offline speech recognition rather then one of the big players interested in my data.
So far i can use rhasspy to switch between radio channels and for start/stop/pause.
With volume i have a problem with using +10 and -10. It works if i put it in my automation but not if i use {{trigger.events.data.val1 |string}} which is +10.
Hello, I tried making this project work as a google assistant actions project only (without dialogflow) but was unable to properly parse the json reply in HA from Google (likely due to my lack of expertise) so in short I found using Dialogflow a simpler (though an extra step in setup) to make work as the Home Assistant Dialogflow addon takes care of parsing the JSON response properly. I may give this option another try when I have some free time. Not likely the near future. Sorry.
Hello All!
I’m having some issues getting this up and running. I’m skipipng the duckDNS setup as i have a static public with my ISP, but under step 13 after you pretty much have it setup, and restart HA, it says you should see a GUI with an LMS control tab and controls. I’m not seeing that, but i’m also not getting any config errors before i restart leading me to believe that i’m doing it right. Does anyone know if LMS controls works with the current release of HA 2022.2.2, I know per the github LMS controls hasn’t been updated in a while, just wanting to make sure this is still current and i should be able to get it working.
Hello everyone,
I’m new to all this and I’ve been trying for hours to make Dialogflow work. I’ve got as far as the google assistant simulator but, I cannot use the simulator and the project never appears in my google home app. All I am shown is the following:
