Fact · Speech recognition
Deepgram's documentation says its diarization recognizes speaker changes and assigns a speaker to each word in the transcript.
Checked against its source on October 17, 2026
In the source's own words
“Diarize recognizes speaker changes and assigns a speaker to each word in the transcript”
Speaker Diarization, Deepgram, n.d., Overview.
Open the original- What it measures
- Word-level speaker tags let an editor split a transcript by speaker after the call, rather than only at the start of each turn.
- What it does not show
- This describes the output format. It does not say how accurate the speaker changes are on a given recording.