Speaker Diarization
Last updated: November 12, 2026
Speaker diarization is available on Standard plan and Premium plan, including the free trial of Standard plan. To enable it, select the Detect Speakers checkbox on the main screen before starting transcription.
The feature is available for all supported media types and can be used with both plain text transcripts and timestamped transcripts. The result is returned as a transcript divided into speaker segments marked with color-coded speaker labels.
Speaker labels can be edited by double-clicking a label. When a speaker label is renamed (for example, changing Speaker 1 to John), all corresponding labels throughout the transcript are updated automatically. This means that to rename a speaker across the entire transcript, it is sufficient to change the label in a single location.