There are three ways to get
speaker ID for your transcripts
and speech analytics

  • Using the channel based speaker ID with stereo audio
  • Providing your own human transcript with speaker markers with the update method
  • Editing the transcript with the player plugin and including speaker markers

In addition to organizing the transcript by speaker, other features like automatic keywords and topics, keyword spotting will include speaker identification.

There are three ways to get
speaker ID for your transcripts
and speech analytics

  • Using the channel based speaker ID with stereo audio
  • Providing your own human transcript with speaker markers with the update method
  • Editing the transcript with the player plugin and including speaker markers

In addition to organizing the transcript by speaker, other features like automatic keywords and topics, keyword spotting will include speaker identification.


See Results

Speaker turns are placed into the transcript JSON similar to words in the transcript. On a speaker turn will increment the word position “p”, start “s” and end times “e” are the start and end times of the current speaker turn in milliseconds, “w” shows the speaker and “m” denotes that this is a speaker turn and not a word in the transcript.