Speaker identification: definition and worked example

In speech science, speaker identification means working out who an unknown voice belongs to by comparing it with several known voice templates.

In speech science, speaker identification means working out who an unknown voice belongs to by comparing it with several known voice templates. In everyday transcription language people also use it loosely for labelling speakers in a transcript. The two are different: labelling needs no knowledge of who anyone is.

Also searched as "speaker recognition" and "speaker id".

What the sources say about speaker identification

  • Wikipedia, Speaker recognition. Speaker recognition identifies a person from voice characteristics; identification compares a voice against multiple templates (1:N), verification checks a claimed identity (1:1); it is distinct from diarisation. Read on 22 October 2026.

Speaker identification in Autrans

Autrans labels speakers with placeholders (Speaker 1, Speaker 2); labels are not names and Autrans does not match a voice to a known person. A person can rename a speaker once and every line they said updates. In meeting minutes, a placeholder such as 'Speaker 2' is counted but never listed as an attendee; attendance is the speakers the owner has named plus the people invited.

A worked example of speaker identification

The transcript shows 'Speaker 2'. The owner renames it 'Amaka Obi'. All Speaker 2 lines now read 'Amaka Obi', and the minutes list her under attendance because she is now a named speaker.

Terms next to speaker identification

  • Speaker diarization: Speaker diarization answers 'who spoke when' in a recording by splitting it into stretches and marking which stretches belong to the same voice.
  • Speaker labels: Speaker labels are the names or tags in a transcript showing who is speaking each line.
  • Voice recognition: Voice recognition is an ambiguous term.
  • Speaker diarization vs speaker identification: Diarization splits a recording by voice and answers 'who spoke when' without knowing who anyone is: Speaker 1, Speaker 2.

Questions people ask

What does speaker identification mean?

In speech science, speaker identification means working out who an unknown voice belongs to by comparing it with several known voice templates. In everyday transcription language people also use it loosely for labelling speakers in a transcript.

How does speaker identification show up in Autrans?

Autrans labels speakers with placeholders (Speaker 1, Speaker 2); labels are not names and Autrans does not match a voice to a known person. A person can rename a speaker once and every line they said updates.

What is the difference between speaker diarization and speaker identification?

Diarization splits a recording by voice and answers 'who spoke when' without knowing who anyone is: Speaker 1, Speaker 2. The full answer, with sources, is on the page for Speaker diarization vs speaker identification.

Where does this definition of speaker identification come from?

It rests on Wikipedia, Speaker recognition, read on 22 October 2026. What each source supports is listed above, next to its link.

Can you give an example of speaker identification?

The transcript shows 'Speaker 2'. The owner renames it 'Amaka Obi'. All Speaker 2 lines now read 'Amaka Obi', and the minutes list her under attendance because she is now a named speaker.

Related

Start transcribing free

Send Autrans into your next meeting, use it in the room, or upload a recording. Your first 30 minutes are free, given once, and no credit card is needed to start.

Last checked 22 October 2026. How we measure accuracy. Spotted a mistake? Write to hello@autrans.online.