Skip to main content

Reference

What is speaker diarisation?

Speaker diarisation is the process of dividing an audio recording into segments and grouping them by who was speaking. It answers "who spoke when". It does not answer "what did they say", which is transcription, or "which named person is this", which is speaker recognition. A meeting transcript with speaker labels is the output of transcription and diarisation together.

What is speaker diarisation?

The three things people mean by "speaker labels"

TaskQuestion it answersOutput
TranscriptionWhat was said?Words, with timestamps
DiarisationWho spoke when?Segments grouped into anonymous speakers
Speaker recognitionWhich known person is this?A name, matched to a voice profile

Diarisation produces "Speaker 1" and "Speaker 2", not "Marta" and "Alex". Attaching real names is a separate step, usually done by matching against the meeting’s participant list rather than by voice, which is why the labels are often right in a two-person call and drift in a six-person one.

How it works

The audio is cut into short windows, typically a second or two. Each window is turned into an embedding, a vector meant to capture the characteristics of the voice rather than the content of the speech. Those embeddings are clustered, and each cluster becomes a speaker. Finally the boundaries are refined, because the initial cut rarely lands where a speaker actually changed.

The number of speakers is usually not known in advance, so the clustering has to decide it. This is the part that goes wrong: given a noisy call, the same person can end up split across two clusters, or two people with similar voices merged into one.

How accurate is it?

The standard measure is diarisation error rate, which adds up the time attributed to the wrong speaker, missed speech, and false alarms, as a proportion of total speech time. Lower is better. On clean, well-recorded meeting audio, current systems land in the low-to-mid teens as a percentage; on the AMI meeting corpus, a standard benchmark with real overlapping conversation, published figures for widely used open models sit around 11%.

What makes it worse

  • Crosstalk. Two voices in the same window produce an embedding that belongs to neither, and overlapping speech is where most of the error lives.
  • Laptop speakers instead of headsets, because the microphone picks up the other participants and the system has to separate a voice from its own echo.
  • Short turns. "Yeah" is a third of a second, which is not much to characterise a voice with.
  • Similar voices. Two people with the same vocal range in the same room is the classic merge case.
  • Low bandwidth, which strips exactly the frequency detail the embeddings depend on.

What to do about it

Very little, at the software layer. The practical fixes are the ones a meeting organiser controls: headsets, one person speaking at a time, and names said out loud early. If a transcript matters, as it does for a hiring panel or a client scope discussion, being able to correct a speaker label once and have it apply to the whole meeting is worth more than another point of benchmark accuracy.

Reference

Questions

What is the difference between diarisation and transcription?

Transcription turns audio into words. Diarisation works out who was speaking during each part of the audio. A speaker-labelled transcript is the result of running both and aligning them.

Can diarisation identify people by name?

Not on its own. It produces anonymous speaker groups. Names come from a separate step, usually matching against the meeting participant list, sometimes from stored voice profiles.

Why does the transcript merge two people into one speaker?

The clustering step decided their voice embeddings were close enough to be the same person. It is most common with similar voices, short turns, and heavy crosstalk.

Try it on your next call

Connect a calendar, let the notetaker join one meeting, and see what comes back. It takes about two minutes and costs nothing.

Free plan available. No card required.