I think I have seen this requested already but can I add my vote to the request to please support ssml.
I use google_cloud and if youi want to use onr of the decent voices it is required. I don’t know if other TTS services support ssml but with Google Cloud it is simply an extra config variable
Using target is a relatively new thing in HA and is always optional - although I have found at least one case using the Services (or Actions - I haven’t upgraded to 2024.08 yet) page that requires it.
HOWEVER
I specifically want to use the form shown in the docs
Were you able to confirm if using the latest beta fixed the issue? I’m still using beta 3 because beta 4 breaks my notifications with alexa devices. (you’re using sonos I see)
I first used Chime TTS when it was first released but to my shame never got around to fully testing it.
I have a stupidly over engineered announcement package that I wrote to do a lot of what Chime TTS does and it has worked well for a many years so inertia took hold.
Chime TTS provides the same functionality and more all in one service call so I am currently migrating all my announcements.
It’s taking me a while as I’m writing a ‘wrapper’ so that I can generate a service call with as many or as few parameters as I like with defaults applied where necessary.
But yes the beta 4 seems stable albeit without any heavy testing yet.
I just hope development carries on with this or at least that @Nimrod_Dolev doesn’t abandon it and that a future HA release doesn’t break it beyond repair. I don’t usually choose to rely on custom components for important things but the advantages of Chime TTS are fairly compelling.
Nimrod_Dolev has been great in addressing issues that have come up! And yeah, I hope that he doesn’t abandon it either.
I think he might be on a break cause we haven’t heard from him as of recent, but that’s okay! Every now and then I walk away from HAss too! We all need breaks now and then. I for one really appreciate the work he’s put into it thus far! If only I knew more about this to help him out, I would.
Hey everyone, I just wanted to update that I’ve been busy with work recently and haven’t had much time for Chime TTS - but I am still here and plan to continue to maintain and extend the integration.
Thanks to everyone for your interest, cooperation and patience
2024-08-14 21:12:32.952 DEBUG (MainThread) [custom_components.chime_tts.helpers.helpers] ╔════════════════════════════════════════════╗
2024-08-14 21:12:32.952 DEBUG (MainThread) [custom_components.chime_tts.helpers.helpers] ║ Chime TTS Say Called. Version v1.1.4-beta4 ║
2024-08-14 21:12:32.952 DEBUG (MainThread) [custom_components.chime_tts.helpers.helpers] ╚════════════════════════════════════════════╝
2024-08-14 21:12:32.952 DEBUG (MainThread) [custom_components.chime_tts.queue_manager] Adding service call to queue
2024-08-14 21:12:33.154 ERROR (MainThread) [custom_components.chime_tts.queue_manager] Service call failed: float() argument must be a string or a real number, not 'list'
2024-08-14 21:12:33.154 DEBUG (MainThread) [custom_components.chime_tts.queue_manager] Queue now empty
2024-08-14 21:12:33.154 ERROR (MainThread) [custom_components.chime_tts] Error calling chime_tts.say service: float() argument must be a string or a real number, not 'list'
2024-08-14 21:12:33.154 ERROR (MainThread) [homeassistant.helpers.script.websocket_api_script] websocket_api script: Error executing script. Error for call_service at pos 1: float() argument must be a string or a real number, not 'list'
2024-08-14 21:12:33.154 ERROR (MainThread) [homeassistant.components.websocket_api.http.connection] [140171008720272] Error handling message: float() argument must be a string or a real number, not 'list' (home_assistant_error) Steve from 192.168.1.63 (Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.0.0 Safari/537.36 Edg/127.0.0.0)
It’s not working for me either. It partially works (the message: Hello) sometimes happens, sometimes not. But the volume doesn’t change any time I run the service.
I’m getting, “Sorry, I’m having trouble accessing your Simon Says NA Skill right now” even though I haven’t changed anything in Alexa Media Player. I can use notify.alexa_media just fine - IOW I don’t get the error like I do with chime_tts.say.
For Alexa users, Alexa Media Player has been having issues lately so I wonder we are still being affected by unknown reasons with AMP.
… EDIT: Looking at the logs again, I see the public URL has an extra ’ / ’ just before “local”.
I checked and I didn’t add 2 of them at the end of my nabucasa url in AMP. So it’s being added by Chime TTS.
Absolutely no offence meant to anyone but it is slightly frustrating that I seem to have hit a block with a very minimal config especially as I am 99% certain it was working before.
I’m now able to play the chimes again via Chime TTS - though no update was done. I think for me it had something to do with Alexa Media Player. There was an update to it today, and now I can play chimes via Chime TTS. I know you’re using Sonos. Just wanted to mention about my situation now.
Hi @Cr4z33! Yes, you can play MP3s but they need to be the following:
MPEG version 2
Bit rate needs to be 48 kbps
Total length cannot be larger than 240 seconds
Chime_TTS can automatically convert any mp3 you save (that’s under 240 secs) in your custom chime audio folder, to the correct parameters using the ‘Audio Conversion’ option shown in Chime_TTS.
Here are two examples (one local file and one internet hosted):
**Note: you may receive an pop up error for timing out, but it doesn’t affect playing the mp3:
Error calling chime_tts.say service: Service call timed out after 20.08s (configured timeout = 20s)
@Nimrod_Dolev, is the timeout something that could be adjusted to match the audio duration of the selected mp3 file so that the error isn’t received? I tried adding a final_delay but it didn’t prevent the pop up error.
Hello, this is not strictly related to Chime TTS, but I was wondering if anyone has any ideas:
I’m trying to tackle a specific use case: I want to fetch an online mp3 audio source and play it. The problem is that the audio is long and takes one or two seconds to load. (The source may be updated anytime and cannot be cached.)
Is there a way to play a chime as the audio source is being loaded, so as to hide the latency?
Something strange since a few days.
In an automation , when a select “action : Chime TTS : say” I have no options.
So my HomePod mini doesn’t “say” anything.
(The problem exist for a new automation but also for old ones which worked fine before)
thanks