You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
When using different TTS voices for primary and fallback, I've noticed that Neon will prefer playing back from a cached fallback response over generating a new primary response.
Steps to Reproduce
Set up a completely different voice as fallback
Break connection to primary server (/etc/hosts blackhole to 127.0.0.1, firewall, etc)
Speak to Neon in a way that will get a consistent response back ("what is the mean airspeed of an unladen swallow")
Reconnect the primary server
Use the same utterance as two steps before
Relevant Code
No response
Other Notes
No response
The text was updated successfully, but these errors were encountered:
Description
When using different TTS voices for primary and fallback, I've noticed that Neon will prefer playing back from a cached fallback response over generating a new primary response.
Steps to Reproduce
Relevant Code
No response
Other Notes
No response
The text was updated successfully, but these errors were encountered: