This is another feature I saw on another automatic transcription service. In my case, it would be useful to be able to specify that there is only one, and enter initials that the transcription would then use at the beginnings of lines. Currently the automatic transcription tries to guess how many speakers there are and ends up with dozens, which just adds to the confusion. My use case: I have separate audio files for each of two speakers, and I can then merge the transcripts together in Nvivo. If I could specify initials for each sound file, the merged transcript would be nice and accurate wrt who is speaking.