I have stereo mikes. I would like Heidi to recognize by voice or by volume one speaker from another and label the transcript by speaker