Using VoiceEncoder as loss with time domain models #84
Description
Is there a way we could use the VoiceEncoder as the loss function with time-domain models?
{{ message }}
Is there a way we could use the VoiceEncoder as the loss function with time-domain models?