Homey notifications spoken by Google Home

Jesper

When somebody opens our backdoor I want a spoken notification that the backdoor has been opened. This can be done by Homey but since all speech is now managed by Google or Alexa it would be better to let Google/Alexa broadcast this notification so that you only have one type of voice in your house for domotica system.

So could Athom create an app that a string of text is converted to speech and then send to the Google Home speaker.

I don't know if this is already possible but I haven't found a way to get this to work. I do have a workaround using tasker and autocast but it would be better to control this directly from Homey.

lubbertkramer

Jesper said:

When somebody opens our backdoor I want a spoken notification that the backdoor has been opened. This can be done by Homey but since all speech is now managed by Google or Alexa it would be better to let Google/Alexa broadcast this notification so that you only have one type of voice in your house for domotica system.

So could Athom create an app that a string of text is converted to speech and then send to the Google Home speaker.

I don't know if this is already possible but I haven't found a way to get this to work. I do have a workaround using tasker and autocast but it would be better to control this directly from Homey.

Good idea but thought it was not possible because of the google/amazon restrictions. If you really would like to know official requests need to be done at the github of Athom as mentioned in the forum rules because this is a community forum

MbenzNo

I did this on my Vera system, but used Text To Speak to my Sonos speakers. (This was before i Got Alexa and Google Home)
A Text To Speak to whatever speaker You have would be nice

Jesper

lubbertkramer said:

Jesper said:

...

Good idea but thought it was not possible because of the google/amazon restrictions. If you really would like to know official requests need to be done at the github of Athom as mentioned in the forum rules because this is a community forum

It should be possible by creating an app which casts a text to speech string to a Google home just like the app auto cast for tasker. Maybe some of the code from that app can be used.

xAPPO

You can now do this on Google Home but not on Amazon Alexa currently.
Google has a ‘broadcast’ feature to all your speakers, or to specific named one.  This will handle TTS. Alternatively you can use the Cast route

swttt

On Alexa the only way i found possible was to use a RPI with bluetooth to stream audio to the echo.
This worked with some latency tho..

Jesper

xAPPO said:

You can now do this on Google Home but not on Amazon Alexa currently.
Google has a ‘broadcast’ feature to all your speakers, or to specific named one.  This will handle TTS. Alternatively you can use the Cast route

True. But how do I initiate a broadcast command using the Homey flow editor?

Roephoed

I did some research on trying to get the output speech from Homey to Alexa (echo dot) and found that this was not possible (hopefully yet?).

My own (tailored for my use case) solution: The simplest way to broadcast speech through the house was to mix both audio outputs in a single audio output with a simple mixer (Around 40 Euro). This works great in my case.

As a bonus, both outputs are now on the multiroom audio system, which Alexa also hears. So with a simple flow I am also able to let homey speak to alexa to issue commands to alexa. (E.g. Homey can say "Alexa, turn on the music' and my Echo Dot starts playing).

Tiyrell

Sounds like a fun solution, having homey issue voice commands to Alexa or Google Home. But  could you explain more about your Mixer and broadcast speech this way? I do not fully understand how this is done at your place.

Thanks!

Roephoed

its as simple as it gets. Both Alexa (echo dot) and Homey have a line out. Both are connected to an audio mixer, which mixes both into 1 line out. Hence your can hear both sounds simultaneous. 

The line out of the mixer is then connected to my sound system, of which 1 of the speakers is close to Alexa. Thus if Homey says something, Alexa is able to hear+understand it. Thus 1) I hear what homey says, and 2) I can automate commands to Alexa (mostly used to turn on music, start playlists etc).

In the end I am happy with this solution, because in this way I tackle both problems at once.