Is there a software that can do this live and could differentiate the voices, so those would get described as Person1 Person2, with timestamp and what they're saying.

  • What platform is this for? Commented May 5 at 22:25
  • MacOS, @RohitGupta Commented May 6 at 23:43

1 Answer 1



Otter has had automatic transcription with multiple speaker identities for a few years.

Import an audio or video file

Audio formats

  • AAC (listed in the Finder with the .m4p filename extension)
  • MP3
  • M4A (including Apple voice memos)
  • WAV
  • WMA
  • OGG
  • but Otter can only do this with a Audio file and not live system audio. Commented May 6 at 23:44
  • 1. Otter is primarily for real-time transcription. So, if "live system audio" means "real time," then Otter can do it. 2. In the contrasting phrase "from system audio and not just mp4 files," mp4 is a container that implies a narrow range of audio codecs. Otter processes file codecs other than mp4-compatible codecs. Commented May 8 at 0:32

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.