Any chance you’ll be adding diarization of known speakers?
For example: from audio samples tagged with names