backchannelingEnabled

This determines whether the model says 'mhmm', 'ahem' etc. while user is speaking.
Default false while in beta.