like i said i dont know if that is possible like that (the fact that the question is there makes me think its not)
the commands for a radio station and for amazon music are different on amazon.
and i dont know if and how its integrated in the component.
but with the IT server it should be possible with the TTS service. (for radio stations that you normally can play with alexa)
I already tried it, but Echo did not spoke what i asked… Simply silence…
I tried again but i got this:
Why the json is invalid?
Yes, exactly, just for example… I have many scripts like this to use with tts service in google home…
Another question: the first time i set up alexa as media_player i did not received the popup asking the passcode… maybe this is the culprit? And if yes, how to be asked again the passcode?
like i said dont use it as tts but to send a command. (TTS wont work on the IT server as you can know if you read back what others did write)
you are confusing stuff.
TTS is text to speech. playing media has nothing to do with TTS.
i dont know if your component is setup correct or not, if you read back then there are conversations about which files to delete.
you are trying service tts.google_say, which has nothing to do with this media player.
you need service mediaplayer.alexa_tts
you cannot use google on an amazon device.
Ok i am really confused… i’ll try to understand it from scratch… Can you tell me which file to delete to have the question about the passcode?
i dont know it without reading back for myself.
so just read back in this topic
/home/homeassistant/.homeassistant/alexa_media.pickle
Just to add on what has already been said, The Alexa does not support TTS, but it is quite easy to have Alexa speak anything you want. Once you get this all up and running, you’ll see that what is happening in the background is that HA is sending a “Simon Says”, to the alexa, which invokes the speech on the echos.
There are many ways to get your echo to speak, most people use a script and then pass what they want to say from automations to the scripts. This is the script that I use, and it works very well.
Probably it does not work in Italian, because “Simon says” is not in the italian Skill list
So if you speak to your alexa and say “simon Says (or the Italian equivalent) - this is a test”, your echo’s don’t, repeat what you’ve said??
Have you tried this
Via the UI , the services tab, choose media_player.alexa_tts :
{“entity_id”:“media_player.YOUR_ECHO_NAME”,“message”:"Hello This is a test’ "}
This video is a few months old, but it looks like it should work. Are you saying that you tried it and it did not work?
If I write
{
“entity_id”:“media_player.echo_dot”,
“message”:"messaggio di test"
}
I get an Invalid JSON error
It looks like you have bad quotes. Probably from the copy/paste. try this
{
"entity_id":"media_player.echo_dot",
"message":"messaggio di test"
}
You’re right, was just a copy/paste problem
In any case this is what I get
It seems you can’t play from the message, you have to download the wav and play it. In any case it does just the sound you hear, one time per Call Service
Yes, I tried it. When one echo says “Alexa,” it wakes the other one but then they both just stop talking/listening.
Sorry, last question as I guess I could just put two of my echos together to test, haha. Thanks for the info
So, if you to a message on alexa A that says “Alexa, what time is it”, the other echo does not wake up and say the time? I realize that’s basically what you said, just asking, since in the video (and a few others), it seemed like with the right combination of words you might be able to get it to work. Now that video may have been before the whole multiple echos in the same area features were rolled out last year.,
thats because in your case alexa interprets your tts as a command.
so try something like: “play some music” (in italian off course)
by the way, why doesnt anyone look at the history from alexa to see what she got???
Well, I guess that that’s either a bug, or the .it servers handle things differently. I personally would not use this if I was not able to easily have my echo’s speak messages. I use that all the time, and it’s great. Having to create wav files sounds like a huge pain.
Sorry I cannot offer any other suggestions.
Good idea about the history, but shouldn’t this
{
"entity_id":"media_player.echo_dot",
"message":"messaggio di test"
}
speak the words “messaggio di test”. Thats how it works for me. If I said “play some music” it should just speak “play some music”