I am testing the Google TTS playback, but have an issue that I do not know how to get around. Currently I have just configured the TTS in configuration.yaml. In the Hass.io gui I click on one of the Chromecast devices, type a text and send it to the speaker. The problem is that the first part of the text is silent. If I immediately after send the same text to the speaker again the whole text is spoken. If I wait like 10 seconds and send the same text the first part of the text is silent.
It is just as if the chromecast needs to be woken up right before the text is sent to the device. Any ideas?
I thought of a similar workaround. How does this SSML tag look like that generates a 1 second pause when us use it ? How would you change the code in the test below? I looked at the break SSML tag, but don’t know how to change below.
service: tts.google_say
entity_id: media_player.floor
data:
message: 'May the Force be with you.'
service: tts.google_say
entity_id: media_player.floor
data:
message: '<speak><break time="1s"/>May the force be with you.</speak>'
But I’m not sure if the Google TTS component supports SSML. I don’t think it does from memory. You’d want to look at Amazon Polly, or one of the other options that supports it.
I had the same problem and I fixed it in Kodi by changing the setting of the Audio.
So in Kodi go to System Parameter --> Audio --> change output configuration to “Fix” instead of Optimize
Thats remove the cutoff at the beggining of TTS command from Home Assistant