All of the pieces for an Amazon Echo (Alexa) <-> Sonos integration.
echo-sonos supports Apple Music, Spotify, Deezer, Deezer Elite, Sonos playlists, Sonos favorites, SiriusXM, the local Sonos music library, and configurable node-sonos-http-api presets.
Sonos Playlists: "Alexa, ask sonos to start playlist MY PLAYLIST in the ROOM"
Sonos Favorites: "Alexa, ask sonos to play favorite MY FAVORITE in the ROOM"
Music services: "Alexa, ask sonos to change music to SERVICE" (SERVICE = Presets, Library, Apple, Spotify, Deezer, or Elite)
SiriusXM: "Alexa, play SiriusXM channel CHANNEL in the ROOM"
SiriusXM: "Alexa, play SiriusXM station STATION in the ROOM"
node-sonos-http-api Presets: "Alexa, ask sonos to play Rock" (queues a node-sonos-http-api macro that sets up speaker/volume/service configuration)
Sonos Rooms and Groups
echo-sonos can default to controlling a specific room, to save you some talking. It will also remember the last room that was used in a normal command, and use that room in future commands.
- Change room: "Alexa, ask sonos to change room to ROOM"
- Change room and service: "Alexa, ask sonos to change room to ROOM and music to SERVICE"
- Add room to the group: "Alexa, ask sonos to join NEW_ROOM to the ROOM"
- Remove room from the group: "Alexa, ask sonos to ungroup ROOM"
- Play songs from an artist: "Alexa, ask sonos to play ARTIST NAME in the ROOM"
- Play songs from an album: "Alexa, ask sonos to play album ALBUM NAME in the ROOM"
- Play a song: "Alexa, ask sonos to play the song SONG NAME in the ROOM"
- Play "radio" songs like this artist: "Alexa, ask sonos to play ARTIST NAME radio in the ROOM"
- Play more "radio" songs like this song: "Alexa, play more songs by this artist in the ROOM"
- Play more "radio" songs like this track: "Alexa, play more songs like this in the ROOM"
- Next: "Alexa, ask sonos go to the next track in the ROOM"
- Previous: "Alexa, ask sonos to go back in the ROOM"
- Clear queue: "Alexa, ask sonos to clear the queue in the ROOM"
- What's playing: "Alexa, ask sonos what's playing in the ROOM"
- Pause: "Alexa, ask sonos to pause in the ROOM"
- Pause all: "Alexa, ask sonos to pause all"
- Resume: "Alexa, ask sonos to resume in the ROOM"
- Resume all: "Alexa, ask sonos to resume all"
- Mute: "Alexa, ask sonos to mute in the ROOM"
- Unmute: "Alexa, ask sonos to unmute in the ROOM"
- Repeat: "Alexa, ask sonos to turn repeat [on,off] in the ROOM"
- Shuffle: "Alexa, ask sonos to turn shuffle [on,off] in the ROOM"
- Crossfade: "Alexa, ask sonos to turn crossfade [on,off] in the ROOM"
- Volume up or down (single room): "Alexa, ask sonos to turn it [up,down] in the ROOM"
- Volume up or down (all in group): "Alexa, ask sonos to turn it [up,down] in the ROOM group"
- Set volume (single room): "Alexa, ask sonos to change the volume to 22 in the ROOM"
- Set volume (all in group): "Alexa, ask sonos to change the volume to 22 in the ROOM group"
Line-In (for Echo Dot)
If you have a default line-in (e.g. it has a dot connected) then you can set defaultLinein in your Lambda environment variables to this ROOM. This will open up:
- Set speaker line-in: "Alexa, ask sonos to set ROOM line-in to the LINEIN_ROOM"
- Play in room: "Alexa, ask sonos to play in ROOM"
This will set ROOM line-in to the default line-in (and play it)
How it works
- When you say the command to Alexa, it triggers the Alexa skill with invocation name "sonos".
- The Alexa skill calls a web service running on AWS Lambda, passing it the preset name ("rock" in the example).
- Lambda then fires an HTTP request to a node.js server running node-sonos-http-api on your local network.
- node-sonos-http-api interprets the command and relays to Sonos over your local network.
Included here are the Alexa API definitions, the Lambda AWS service that catches the Alexa requests, and an example preset configuration for jishi's node-sonos-http-api to actually play the music.
To set it up, you need to do the following:
Get jishi's node-sonos-http-api working
- Install node.js on a server on the same network as your Sonos. By "server", we mean any computer that will never be turned off or put to sleep. If you're having trouble on Windows, try this blog. On Mac, check your version of node with "node -v". If it's less than version 4, you need to upgrade node.
- Download node-sonos-http-api. The easiest way to do this is to open a command prompt (Terminal on Mac, or type "cmd" in the Run menu on Windows) and type "npm install https://github.com/jishi/node-sonos-http-api". If you downloaded it as a .zip file, unzip it and put it somewhere easy to get to in a terminal window (like right off C:\ on Windows or in ~ on Mac).
- Take the node-sonos-http-api/presets.json that I have here and drop it into the "presets" folder in your node-sonos-http-api root directory. Modify it to use your speaker names and your favorite stations. Make sure the preset names are lowercase (like "test" and "rock" in my example). NOTE: You can skip this step if you only want to use Playlists, Favorites, or Advanced Mode, which require no configuration.
- In your command prompt / terminal, go into the node-sonos-http-api directory ("cd node-sonos-http-api") and type "npm start". You should see a bunch of stuff indicating that it's running now. If it dumps you right back to the command prompt, something went wrong - likely you have some invalid characters in presets.json. Try opening presets.json in a text editor, copying everything to clipboard, and pasting it in a JSON syntax validator. Fix whatever errors it finds, paste it back in the file & save it, then "npm start" again until it works.
- Test it by hitting http://yourserverip:5005/zones
- If you get a response, great! You now have a server that can control your Sonos. Now try playing something: http://yourserverip:5005/preset/[your_preset_name]. Or, play a Playlist or Favorite (example: http://yourserverip:5005/kitchen/playlist/myplaylist). To stop, use /pauseall.
- If you have problems, make sure you are on the same network as your Sonos AND make sure you don't have a Sonos client running on the same machine. The client can interfere with the node.js server.
- Optional: You probably want to set this up to auto-start in the event it dies. Auto-launch is one of many solutions to make that happen.
Expose your server to the outside world
- You need some way to make your server accessible from outside your home, so Alexa can contact it.
- In most homes, your internet address changes over time based on however your internet provider does things. We need to come up with a static address that we can tell Alexa, so your server isn't a moving target. Services like yDNS.eu and DynDNS will do this for you (and have setup instructions for your platform). Your router may also have a feature called "Dynamic DNS", which is the magic term for what we're trying to do here. Asus routers have this, for example. You go in and configure a name for Dynamic DNS, like "bobspalace", and it will give you a name like "bobspalace.asuscomm.com".
- Once you've figured out the whole permanent hostname thing, Alexa knows where you are. However, your router doesn't know you have a server running, so if it sees connections from Alexa, it'll drop them because it doesn't know where to send them. So, next we need to setup something called "Port Forwarding". When your router sees an incoming connection to a certain port (5005), it will forward that connection to your server.
- To setup Port Forwarding, we have to first figure out the IP address of the server we're on. Here's a solid guide for Windows, and here's one for Mac. (Note: This IP address may change over time, too, if you reboot this computer or your router. There is a place in your router configuration where you can setup "Static IP addresses". I recommend you set one up for this IP address, so your router will always give this computer the same address.)
- Next, find the "Port Forwarding" section in your router's admin console. Put "5005" as the source port and the destination port, and for IP address, use the one that you took from step 4.
- Finally, we check that it worked. Test it by hitting http://yourdyndnsaddress:5005/zones. If you get a response, then great! Your dynamic DNS address worked and your router properly forwarded 5005 to your server.
Create the Alexa Skill that will send events to AWS Lambda
- Create a new Skill in the Alexa Skills control panel on Amazon. You need a developer account to do this. The account must be the same as bound to your Echo, and make sure you are logged into that account on amazon.com. You will see an error indicating access denied if the two accounts are different.
- Name can be whatever you want. "Invocation" is what you say, and it must be all lowercase (I used "sonos").
- Check Custom Interaction Model if it is not already checked. Click Next
- Click Next, taking you to Interaction Model. Create a Custom Slot Type ("Add Slot Type"). Add a new types for PRESETS, ROOMS, SXMCHANNELS, SXMSTATIONS, SERVICES, TOGGLES, ENDINGS and a final one for NAMES. Into each, copy/paste the contents of echo/custom_slots/PRESETS.slot.txt, echo/custom_slots/ROOMS.slot.txt, echo/custom_slots/SXMCHANNELS.slot.txt,
echo/custom_slots/TOGGLES.slot.txt, echo/custom_slots/ENDINGS.slot.txt. and
- Still in Interaction Model, copy this repo's echo/intents.json into the "Intent Schema" field, and echo/utterances.txt into "Sample Utterances". Note: if you're in the UK or Germany, Amazon's support for band names and songs may not be available yet. In that case, replace anywhere you see
"AMAZON.MusicAlbum" with the word
- Don't test yet, just save. Click back to "Skill Information" and copy the "Application ID". You'll need this for Lambda.
Configure the AWS Lambda service that will trigger your node-sonos-http-api server
Create an AWS Lambda account if you don't have one already. It's free!
In the Lambda console, look to the upper right. Make sure "N. Virginia" or one of the Lambda supported regions is selected, because not every zone supports Alexa yet.
Create a new Lambda function. Skip the blueprint.
Pick any name you want, and choose runtime Node.js.
Go into this repo's lambda/src directory and copy options.example.js to options.js. It's recommended you DO NOT edit the options.js file (leave all defaults), because it's easier to customize this later.
up everything. On Mac/Linux,
cd src; chmod a+r *.js sonosProxy/*.js; zip src.zip *.js sonosProxy/*.js. Make sure you don't capture the folder, just the files.
Choose to upload the zip file for src.zip.
In the environment variables section make sure to fill in the following environment variable
APPID - this is the Application ID you copied from above
HOST - this is the DNS name for your server hosting node-sonos-http-api
To enable music services and default rooms (strongly recommended), also add the following environment variables
true) - remembers last room you commanded, and uses your default music service
DEFAULT_MUSIC_SERVICE - supports presets, library, apple, spotify, deezer, or elite
DEFAULT_ROOM - example: "Kitchen". If Alexa doesn't remember your last room and you don't say one
You may optionally fill in any of the following environment variables
AUTH_PASSWORD - for basic auth with node-sonos-http-api
false) - if you enabled https on node-sonos-http-api
false) if you are using node-sqs-proxy for secure communications
The default handler is fine. Create a new role of type Basic Execution Role. Pick smallest options across the board, because this is the world's smallest service (smallest option for memory, and so on). Set the timeout to something high (suggest 30 seconds or so). For everything else, the default options should be fine.
Click Next to proceed. Once created, click "Triggers".
Add a source. Choose "Alexa Skills Kit".
Test it out. I included a test blueprint in this repo. Click "Test" and copy/paste this repo's lambda/play_intent_testreq.json to test. It will trigger the "test" preset in your presets.json file on your Sonos server. Don't forget to replace the Alexa App Id again.
- You should see the following output if successful:
For Advanced Mode you also need to give Lambda permission to access DynamoDB for storing the current room and service settings. Simply click on your AWS Dashboard account name in the upper right corner, Security Credentials, close the popup message, click Roles on the left, click on lambda_basic_execution, click Attach Policy, click the checkboxes next to AmazonDynamoDBFullAccess and AmazonSQSFullAccess, and click Attach Policy at the bottom of the screen.
Connect Alexa Skill to AWS Lambda
- In the Lambda console, copy the long "ARN" string in the upper right.
- Go back into the Alexa Skill console, open your skill, click "Skill Information", choose Lambda ARN and paste that ARN string in.
- Now you're ready to put it all together. Try "Alexa, ask sonos to play test"
Optional Security Features
The echo-sqs-proxy solution allows Echo-Sonos to communicate with the node-sonos-http-api solution without having to alter your firewall to open your server to the Internet or having to make any of the changes below. Read the README file in the echo-sqs-proxy directory for instructions.
Alternatively, if you're not using echo-sqs-proxy, echo-sonos does support both HTTPS and basic auth. options.example.js has flags and configuration information for both features. For HTTPS support, set "useHttps" to "true" and check to make sure the port is still correct. For basic auth, change the "auth" variable and replace the username and password with your own.
Securing node-sonos-http-api, HTTPS, and your home server are outside the bounds of this project. The below documentation is provided for convenience because so many people have asked me about it. You could certainly do much more to secure your home server. For starters, you could pin certificates to the client or put more effort behind secure key and credential storage. This is a DIY hobbyist project, and it's up to your discretion to determine how much effort to put into security.
Both HTTPS and basic auth need to be configured on node-sonos-http-api before echo-sonos can use them. You can configure one without the other, but they are recommended together. In your node-sonos-http-api directory, create a file called settings.json according to the node-sonos-http-api documentation.
An example might look like this:
In the above example, HTTPS is configured using "yourserver.key" and "yourserver.crt". For utmost security, it's best to purchase a certificate from a reputable certificate authority such as Digicert. You could also use a free one from a service like LetsEncrypt.org. A final option for those who know what they're doing is to self-sign, and if you choose this route, then you must set "rejectUnauthorized" to "false" in options.js.
- If you have trouble with your node server not triggering the music even when you hit it on localhost, it probably can't find Sonos. If it crashes with a message about "discovery" being "null" then that's definitely the case. Usually you're on the wrong wifi network, you forgot to close your Sonos application (which screws everything up), or your server died for some reason and you didn't notice.
- If your Lambda test doesn't work, then you might have a case mis-match between the preset name in presets.json and the value in the Lambda test. It's also possible Lambda can't reach your host because your DynDNS setup isn't working, or a firewall is blocking it. If you're unsure, try the Maker plugin on IFTTT, to see if you can get it working externally from someplace else.
- If Alexa says something about not being able to use your skill, then Lambda is probably returning an error. Check the Lambda logs. Alexa will say this if she doesn't see a proper response from the Lambda server.
- If you run into a syntax error on node-sonos-http-api that looks something like "Syntax error in module 'index': SyntaxError at exports.runInThisContext", then it's likely that you inadvertently edited presets.json with a rich text editor and it replaced some of your quotation marks with quotes from a weird character set. Try pasting your presets.json into a JSON linter like jsonlint.com and it should point out this error.
- If you have trouble with Alexa recognizing commands, make sure you say "Alexa ask sonos" rather than some of the other possibilities (like "Alexa tell sonos"). Many people report that "ask" works better than "tell", and (strangely) that a lowercase invocation name ("sonos") is better than its camelcase variant ("Sonos").
If you're still stuck, add a comment on my original Amazon Echo + Sonos integration blog post and I'll try to help you out.
When upgrading your code to the latest version, make sure you do the following:
- In the Interaction Model under the Alexa Skills Kit console, update the Intents, the Utterances, and the two Custom Slot Types
- Zip all of the .js files (without the folder - just the .js) and update them in Lambda
Lots of people are forking echo-sonos, which is awesome. I'd love to bring some of that innovation back into the project, so don't be shy about submitting pull requests!
Thanks to @jishi for node-sonos-http-api, without which none of this would be possible. And major credit goes to @jplourde5 for a slew of big features, like music services and echo-sqs-proxy.