Closed ken-mycroft closed 2 years ago
One thing to note is that we might also need update the handling of the tts-cache. I think in this setup only one of the TTS's cache clearing methods will only be called...
I was hoping we could merge into both branches, however I've been testing this out on the standard dev channel as well and I think we must have some extra fixes to the mimic1 class in feature/mark-2. If you set your TTS module to mimic it starts playing all sentences in the queue at the same time.
Merging this now so we can get it onto tonights build but agree with the extra points raised so we should address those as a followup.
Ok interesting, I'll try that out and see if I can make this work for dev aswell.
I was worried about this. There might be some mimic1/2 changes i missed. Is this still an issue?
I added a couple of things in the backport to dev PR #3055, it has commit for handling the cache clearing and some minor reorganization moving responsibility to the playback-thread. It also has some further fixes where the fallback could misbehave
Description
Fixes issue where when remote tts fails and local gets executed they stomp on each other
How to test
You would need a special mimic2_tts.py that periodically throws exceptions, otherwise not easy to test
Contributor license agreement signed?
CLA [ yes ] (Whether you have signed a CLA - Contributor Licensing Agreement