Yep. @jg506 was gracious to give me tips and suggestions for getting this to run on my Synology via Docker. Everything is awesome and I love the announcements.
For anyone reading this, if you’ve found a way to change the voice, I’d love to explore how to do that. I assume it would be configured within the device handler, but not sure if @vervallsweg has that in his code.
set up a docker server, and there are several instances that you can just download and they work immediately. I have a Qnap Nas with container station. I searched on a whim for cast-web and got two results back. installed the first one, and set a static IP to the virtual device, and was literally running announcements through my GH 8 minutes after searching for it.
@vervallsweg,
I’ve had everything up and running for some time now (thank you!). When I attempt to add a new Google Home Mini via the SmartApp, it seems to be recognized by the discovery process but clicking Done gives me “Error saving unnamed page. Please contact SmartApp developer.” I’ve seen this issue before (when I was transferring the API from my Windows computer to Docker) and the solution I followed was just deleting all my devices then re-associating them. Is this the appropriate workaround or is there something else to do when this happens?
UPDATE: Apparently, removing all my devices from the app (and confirming via IDE) didn’t actually help get everything re-associated so now I can’t add anything back in. I ended up deleting the Cast web - service manager SmartApp then back in and I was able to associate everything. I did notice throughout the process that the SmartApp was providing results with no values (just empty lines with a checkbox next to them) so I wonder if that’s a defect in the SmartApp.
Also, when on the “Select Cast Device” screen, is it possible to put the device’s name instead of the Device Network ID value? Not sure if you are able to get a value from the device itself or perhaps whatever it had been labeled if it was already associated to ST as a Thing.
Awesome job @vervallsweg !!
Been running it since a day or two and still trying to make the “PREV” and “NEXT” work on spotify.
Anyone had any success ?
Also is it possibile to use the presets with spotify playlists ?
Thanks !
Any suggestions for what to do when devices are missing? I’ve installed everything today and it found one of my google home devices but not the other, nor my nvidia shield. Both are linked to my main GH. I can work with just one, but I would prefer to be able to get notifications on different floors
Hi… i see you installed this on a pi. I believe i was able to clone the github but i cant get npm install to work. I made sure i was in the right directory.
Edit: Found the help updating node.js. That did the trick.
Using this with Plex server. Both running on a PI 3. So far very COOL!!!
Thanks
You can try manually increasing the discovery time, could be related to this ‘bug’. Just realized that I still didn’t find any time to document that parameter, 
for the meantime this issue should help you to change it.
Well this is an interesting one to say the least 
. The problem is roughly documented in this GitHub issue.
TLDR: the cast protocol supports it, but I have never seen any cast application ‘out in the wild’ that supports it. Spotify and even Google’s own play music all don’t support skipping or reversing media.
As long as there’s no receiver that supports it I didn’t even bother implementing it and instead give you the option to remap those buttons.
No, but has been discussed here already.
If I remember correctly somebody already reported a similar issue that I think magically fixed itself. Will update if i find it.
Well this error message also appeared to me at some point and I couldn’t find anything relevant in the logs that would tell me what the problem is.
What do you mean by that? Did you remove all the devices and smartapps via IDE and after reinstall they were still there? Maybe let me know which version of the API, DTH and smartapp you’re running?
ST TTS, the thing that I’m using is barely documented. If somebody tells me how to change the voice from ‘polly’ to somebody else, I’d be glad to do it. However I doubt that this should be implemented in the DTH, probably in the smartapp itself. But I’ll look into the smartapp someone posted here that offers more TTS choices 
Update:
Just looked into it and apparently there are more online TTS services that are somewhat free. However I don’t think this belongs into the DTH. Maybe I’ll write a separate universal Smartapp that works with every music-player. Or maybe even the author of the DLNA dth can unbundle the third party TTS implementation.
Can you run the API with debugging enabled? Then run /getDevices again and post the output here.
If I may ask, where did you get these instructions from. They’re quite outdated and I should have removed them already.
Good one 
. Well the problem is that njs is so diverse and runs on so many platforms that I cannot compile a guide for every platform here. And if I was to focus on the mayor platforms, imho there are already tons of good instructions out there on how to get njs running.
But I understand how frustrating that can be and also wish it was easier.
Yes and the approach you took looks good. Also possible through webCoRE if you’re into that.
Looks alright, but no “” around the addresses, etc. and make sure to url encode everything that has a space, etc in it. If you paste that into Chrome, it should automatically do that, otherwise there are online tools for url encoding.
No, but doing so consecutively should work, 1. device volme 2. playback
If you still have that error please give me the API version and enable debugging, just like 177
What exactly are you trying to play?
Is no valid stream type, see documentation for reference.
That means it’s broken. Just post it here and I can try to fix it.
No. You can only paste them just like in the mobile app.
Ahm… there should be a button under the mobile app text field… which version of the DTH are you running?
And which has been working perfectly with ActionTiles!
Yes, I realized that, and used a URL encoder to parse the station names and subtitles into URL-formatted strings.
Then it’s already choosen or you have just one installed. Try installing nano on your system, the most usable editor and then add it to the bottom 
Update: Here’s the command I’m using
@reboot sleep 30s; forever start cast-web-api/castWebApi.js --hostname=192.168.x.x
Forever also supports execution as daemon directly, just haven’t looked into that yet.
Yes it is. Either through speaker companion, webCoRE or basically any other smartapp.
The biggest hurdle that I had was getting the correct version of node installed. Because I was running Dasher on the same Raspberry Pi, I have v4 of Node installed instead of v6. That’s why the installation was failing. It seems that quite a few other people were running into the same thing. I posted the instructions on getting the proper version of Node installed. Including those might be helpful.
Pi3:
1. wget https://nodejs.org/dist/v6.11.5/node-v6.11.5-linux-armv7l.tar.gz
2. tar -xzf node-v6.11.5-linux-armv7l.tar.gz
3. cd node-v6.11.5-linux-armv7l/
4. sudo cp -R * /usr/local/
pi Zero/ZeroW:
1. wget https://nodejs.org/download/release/v6.11.5/node-v6.11.5-linux-armv6l.tar.gz
2. tar -xzf node-v6.11.5-linux-armv6l.tar.gz
3. cd node-v6.11.5-linux-armv6l
4. sudo cp -R * /usr/local/