I opened an issue for the different parameter type when using the different versions of the say action.
If it is accepted, I will propose a fix to expect a String in all cases.
The “funny” thing about that is, the parameter for the text is always expected as a string, at least the code looks the same for me!
The only difference migth be, that while a single string is expected, an item.state will automatically be understood as a string, while in case of other overloads, it is not.
It is now fixed.
In few days, the snapshot will no more require the .toString() whatever the say or interpret actions wihen using an item state as parameter.
The fix should be included in the snapshot 623.
@Marty56: did you create an issue? I couldn’t find one, but i’m not (yet) that familiar with github and co.
I experience wrong sample rates at both german languages, only english does work at my configuration with the simple “say” command (changing the voices through paper UI)
Please mention the TTS engine you are using.
No problem for me with Voice RSS.
Yes, with Voicerss it is working nicely. The disadvantage of this approach is that it is a cloud service.
So I would still prefer marrytts.
So open an issue for marytts in this case.
already done for some time.
can you link the issue here? i’m to stupid to find it…
three post above Kai already provided the link but for your convenience here it is
In build 644 I have tested the newly added support for the spdif interface of the playbar in combination with say and playsound.
I was expecting that the say command would be switch volume to the notificationvolume, then play the text and switch back to the spdif interface.
This is not working. In logs I am receiving no error but it seems that the TV input is not changed.
Doing the same playing music is working nicely.
I am not sure whether I have to add any further command e.g. to turn on TV on or off or whether it is just a bug.
I am sorry, I tried to implement something while I have no playbar and so no way to test it. I probably missed something.
I appreciate you trying though not having a personal benefit.
Sorry that I could not support your development better in the PR.
644 looks like an old snapshot ! It probably does not include my patch. Please try with a recent snapshot.
That was a typo it is build 664
I have a question regarding volumes. I set everything up and tested with the following rule:
rule "Test"
when
Item RuleTest received update ON
then
setMasterVolume(new PercentType(50))
say("Hello Everybody")
end
Everything worked great. The problem is I would like to return the volume back to the volume from before it was adjusted. So for example, I’m listing to music low in the background. I’d like to raise the volume, make an announcement, and then continue the music at the original volume. I assume this is possible I’m just not skilled enough to figure it out.
I’d try getMasterVolume ().
What kind of audiosink are you using?
If it is Sonos, that switching back of volume isn’t needed because on Sonos devices the say command uses the notification volume and not the volume.
It is Sonos. How do you change/set the notification volume? I didn’t see anything about it in the binding documentation.
cut and paste from documentation
notificationvolume Dimmer Set the volume applied to a notification sound all