rt,We can identify a speaker that have benen trained,how about a new speaker's voice?Can it be identify?