HowTo: Listen & Talk to your Home

[H102 (CM6.5 H102)]

Excellent tutorial @Dim, keep up the fine work and Thank You!

[rlkoshak (Rich Koshak)]

Fantastic tutorial! I know I’ve always found TTS to be a bit of a mystery. Thanks for posting!

[Dim (Angelos)]

@rlkoshak

I know I’ve always found TTS to be a bit of a mystery.

It’s funny: I was also worried about this, but it was easy to setup. You need to be careful with Google’s Cloud Platform and to perform the steps in the docs to get the JSON. After that, it’s very smooth.

This will be improved according to my understanding, since @wborn is planning to change the API (Ref: Google TTS addon for OH2 - #63 by wborn & https://github.com/openhab/openhab2-addons/pull/4221). I didn’t use the testing jar for the tutorial since I would have pushed everyone to Snapshots to use it

[Kim_Andersen]

Damn fine tutorial @Dim
But, I seem to recall something about Google text to speech doesn´t work on Rpi. If thats correct, perhaps it´s worth mentioning in your guide,

[Kim_Andersen]

@Dim About your FAQ Q3:
Google Assistant (and openhab) also support [ “Outlet” ] and [ “CurrentHumidity” ].

[Dim (Angelos)]

@Kim_Andersen

If thats correct, perhaps it´s worth mentioning in your guide,

It’s correct. I added a note on top (intro). There is a Snapshot release of the addon that will work on rPis, but you need to run 2.4.0 Snapshot core to run it. Ref: Google TTS addon for OH2 - #73 by wborn

@Kim_Andersen

also support [ “Outlet” ] and [ “CurrentHumidity” ].

Check (updated). The docs need update also in this case

Thanx!

[Kim_Andersen]

@Dim

It’s correct. I added a note on top (intro). There is a Snapshot release of the addon that will work on rPis, but you need to run 2.4.0 Snapshot core to run it. Ref: Google TTS addon for OH2

Great… I really look forward to skip the Voice_RSS. I want Google both ways

@Dim

Check (updated). The docs need update also in this case

This is unfortunate yes… This page is more updated that the original is
https://github.com/openhab/openhab-google-assistant/blob/master/USAGE.md

[Dim (Angelos)]

added 2 more FAQs. I think that they are useful in tutorials

[Dim (Angelos)]

@Kim_Andersen

This page is more updated that the original is

aha! the openhab documentation bot is slow again
It should have pulled that version and published it on Introduction | openHAB by now (last update to your link is 26 days ago

Thanx for the link ! I will further update the supported tags

[Kim_Andersen]

Regarding the [“CurrentHumidty”] there is a small note to catch. It doesn´t show anywhere. It´s part of the thermostat. It confused me some days ago, when I added it to my setup. One simply have to ask for the humidity to know if it works.
Since this guide is is about listen and talk to. I´ll let you decide if it´s worth mentioning

[H102 (CM6.5 H102)]

Here’s one you might want to include on the Q&A, or at the top of the topic.

Q: How can I get this to work with Alexa?

A: This tutorial is for Google, not Alexa.

Surprised no ones asked this yet.

[Dim (Angelos)]

@H102

A: This tutorial is for Google, not Alexa.

Added

[H102 (CM6.5 H102)]

@zamzon

Is it true that I can’t use the Google TTS service on my Raspberry pi 3

I get this error when I try to install the add-on

Please read this, also located at the end of the tutorial.

@Dim

Please do not use this thread for troubleshooting purposes!
If you are having trouble setting up any of the above, please refer to the respective sections in the forum and search for existing solutions. If you cannot find one that solves your specific problem, create a new thread linking to this tutorial and adding as much info as possible about your setup!

Thanks

[Kim_Andersen]

@Dim

2.2.5 Configure the Default Sink to be Web Audio under PaperUI -> Configuration -> System -> Audio

Hi Dim…
I have a question. (well, partly an request ).
My plans is to use several devices for “listen and talk to” spred around all over our house. That include several Google home devices, as well as one android tablet, (main communication device). I have however yet to discover how to use a tablet for both talk to as well as listen…
As far as I understand you´r tutorial, you´re dealing with one device only, which is the default sink? I wonder if you could spred some light about this issue of using more than one device, if that is possible at all?

[Dim (Angelos)]

The default sink in my tutorial is Web Audio. This means: Any Web Browser pointing to either PaperUI or HABpanel URLs.

I used 2 devices: my PC and my mobile to test things out and I had PaperUI always open on the PC, so I was able to listen to the TTS audio.
If you have 2 browsers (or tabs) open to both PaperUI and HABpanel, you will listen to the same audio output twice

I don’t know if you use a tablet and have the browser in the background, working with the Google Home App in the foreground, if you will be able to receive the Web Audio (most likely no ).

The audio feedback is “channeled” only to the browser (not to the GH app)

Since you will have Google Home devices available, you can send the TTS Audio to them (either configure the default to Sink to be the GH device or specify the audio sink in your say command)

[Kim_Andersen]

@Dim

If you have 2 browsers (or tabs) open to both PaperUI and HABpanel, you will listen to the same audio output twice

Thats smart Except, in my plans…

@Dim

I don’t know if you use a tablet and have the browser in the background, working with the Google Home App in the foreground, if you will be able to receive the Web Audio (most likely no ).

Since you will have Google Home devices available, you can send the TTS Audio to them (either configure the default to Sink to be the GH device or specify the audio sink in your say command)

… is a bit tricky, cause I will not be using web audio. I will be using chromecast´s devices as “speak to”, as well as “listen to” TTS audio. And even more, I plan on making somekind of presence detection, and only send TTS audio to that/these rooms where presence is detected. I know, this will require rather advance rules. But at least thats an idea.

I know how to cast TTS audio to chromecast devices. I already do that with the one Google Home Mini I´ve got. My concern is, how to cast TTS audio to an Android tablet. I dont even know if thats possible, unless I´ll be using the web audio. I guess this is a question to be searched for…

[wborn (Wouter Born)]

@Dim

Important Note: The Google TTS Add-on does not work currently on Raspberry Pis due to some dependencies. This will be improved soon and we expect openHAB 2.4.0 Stable Release to include the new version that will work on ARM architecture also.

It should now work out of the box on Raspberry Pis with 2.4.0-SNAPSHOT Build #1446 or newer.

[Dim (Angelos)]

@wborn

It should now work out of the box on Raspberry Pis with 2.4.0-SNAPSHOT Build #1446 or newer.

Thanx! I am updating the first post now

Quick clarification question: Everything with regards to the installation instructions posted on: Google Cloud Text-to-Speech - Voices | openHAB remains the same… Correct? (get JSON, plug it in PaperUI)

[wborn (Wouter Born)]

@Dim

Correct? (get JSON, plug it in PaperUI)

That’s right! It works exactly the same on any architecture while the addon got a lot smaller.

[Dim (Angelos)]

this thread has started to show up on https://www.openhab.org/about/showcase.html

It also has 6k views on twitter https://twitter.com/Angelos_F/status/1064609328156631043

We need to improve the content on the first post… what about making it a Wiki?