Echo Devices (Alexa) as Media Player - Testers Needed

Thanks. Buit is there a way to get the string spoken to that alexa

I think just the “alexa_tts” service will… I started switching all my stuff over to the notify.alexa_media platform with announce last night… the beep before it speaks is exactly what I always wanted so you can start listening before she speaks. Working great! Awesome work guys.

as stated above the only thing being deprecated is the media_player.alexa_tts service.

It will be switched to the notify.alexa_media service and the data : type: will be “tts” and it will still work the same as it does now.

1 Like

The service looks right. Can you enable debugging for alexapy? You can review the actual command sent with announce which should list all 5 echos when it sends the announce. It probably makes sense to move over to Github to debug further or you can DM me.

Yes, thanks for helping to spread the word @finity . To reiterate, media_player.alexa_tts service will be gone in 1.3.0. Use notify moving forward.

What string? Review the automations people have posted and see if any work for you.

Hello !

I was using the component in 0.88, and I updated to 0.89. So, I replaced the old component with the new version du to the great migration. And now, it can’t log to my Amazon account. It ask me to enter sms or email to get a 2FA code, but when I do so, nothing append. The configurator get stuck on this : https://img.wazo.ovh/9BL2XCB

Any solution ?

EDIT : The solution was to enable 2FA on my amazon account !

1.2.2 is out.

This is mainly a maintenance release with some improvements from websockets. NOTE: If you receive websocket_client errors, the version on your system may need to be updated to 0.55.0. We’ll pin it in the next release.

Fixes:

  • Fix bug where emulated_hue and other smart home device routines failed. (resolves #123)
  • Fix bug where Alexa devices that came online later were not properly updated. (resolves #125)

Changes:

  • Add websocket support (resolves #119) - This allows us to remove most polling and rely on push notifications from Amazon. This also means that media_players should update regardless of how or where you interact with them as last_called should update immediately. As a result alexa_media.update_last_called service may no longer be required. Prior polling methods remain as a fallback mechanism if websocket can’t connect which may be based on region. No additional configuration is needed.

Full changelog .

As always, please use the the Wiki and FAQ if you need instructions or have questions. Also, please help in documenting features in the wiki since this thread is basically unsearchable now. We also use GitHub for bug reports/feature requests .

4 Likes

@alandtse How can we proof if it’s working without alexa_media.update_last_called.

Hello,

On the last update of Hassio 0.89, do I have to put this on the config.yaml ?

api:
alexa_media:
accounts:
- email: login
password: pass
url: amazon.fr

What file do I have to put on the /config/custom_components/ ? (I understood the alexa.py is not required anymore).

Thanks for the help

did you look at the installation instructions?

also, make sure you correctly format your code that you post so people can see errors in syntax.

I just tried a couple of quick tests not using update_last_called and seems to work fine now without it. :slight_smile:

1 Like

FYI, we found a bug for those who are able to get websockets running initially. It eventually times out and doesn’t degrade back to polling. We are working to resolve it. When the bug hits, you’ll notice last_called stops updating immediately unless you use the alexa_media.update_last_called service and the media_player may stop updating while playing.

However, if you received a error saying websocket failed on startup, then you don’t have this bug since you never got websocket to start and the polling fallback does work.

I have that error. So what does that mean? This new feature looks primising. Should we wait for the next version for this to fixed? Or does it work and we can ignore the error.
If it’s not neccesarry I would like to comment out alexa_media.update_last_called from my automations.

Also some of you wrote about the setting an announce group within the alexa App I currently don’t know if I can’t find it or if it’s not available for my region. Where can I find this feature?

Salut,
I had problems with amazon.fr try amazon.com

can someone add French to “Known working languages for TTS” ?

1 Like

Hi everyone, I’m sure I followed everything here but need some help troubleshooting 1 thing.
The component loads, all my 5 Alexa devices are recognised. However, there all have “unavailable” next to them .

If have tried running this to no success.

Service
media_player.alexa_tts

{ “entity_id”: “media_player.raymond_s_echo” , “message”: “Back door is open”}

Any advice? Help please?

Thanks

Has anyone come up with any way to play a custom track? Apple music, spotify, amazon music. I’d like to upload a recording of my own voice to play. I have an automation that uses one echo to tell another echo what to do. It works, but only about 30 % of the time. I’ve tried so many variations, and have just given up. If I could play my own recording, I suspect that would work

What do you want to tell her, maybe there is another way to do this.

I’m using it as a hack to run another Alexa skill tied to my car. Alexa, tell blue link to lock my car.

Then wait for a response then provide a 4 digit pin.

So i have two echos near each other. One gives the command and the other runs the skill

Works sometimes but not consistently

If you see the websocket error on startup, you can ignore the error and use 1.2.2 without any issues. We fixed the debugging message to make it clear we fallback to polling. Keep your alexa_media.update_last_called in your automations for now since it’s needed for polling.

I think people are referring to creating Alexa Multi-Room Groups and using Announce for that group. The issue is you can only have an Alexa device in a single group.

Ensure your Alexa domain is appropriate for your region. For example, if you’re in Germany, you need to be using amazon.de and not amazon.com. If you’re logged into the wrong domain, none of the devices will show up as available.

The component will only work with Alexa devices that are available.

1 Like

You mean if it’s a normal group and not a multiroom group, right?

1 Like

I guess a better question would be. With Apple Music or Spotify (not sure about amazon prime, but not the paid unlimited), can I play a track from my library. For example I have several live tracks of concerts in my iTunes library. Can I play those? If so, it seems like it would be easy enough to upload a track of me speaking.

Has anyone used this component to start a track from their library that was uploaded to the cloud.