LoginTry it Free

What Are Transcription Conventions?

Transcription conventions are the agreed rules for turning speech into text: how to mark speakers, pauses, overlapping talk, emphasis, laughter and words you cannot make out. A study or a project picks one set of conventions and applies it to every transcript, so that transcripts made by different people can be compared.

Example
Jefferson:
A: so you (.) left in ma:rch?
B: [yes ]
A: [right] and THEN (0.8) nothing

Simple rules:
P04 [00:12:40]: So you left in March?
P02 [00:12:42]: Yes. [laughs]

Jefferson notation

Developed for conversation analysis, it marks how something was said, not only what: (.) for a micropause, (0.5) for a timed pause in seconds, [ and ] where overlapping talk starts and ends, = for speech that follows without a gap, a colon for a stretched sound, CAPITALS for louder speech and degree signs around quieter speech.

GAT 2 and other detailed systems

Discourse analysts, especially in German-speaking research, often use GAT 2, which marks pauses, accents and intonation at several levels of detail. Like Jefferson, it is chosen when the way people speak is part of what is studied.

Simple transcription rules

Most interview studies use a light set of rules: speaker labels or participant codes, a timestamp at each turn, clean verbatim text, and a few markers such as [laughs], [pause] or [inaudible 00:12:40]. They are quick to apply and easy to read, which suits thematic analysis.

Questions

Which transcription conventions should I use?

The ones your method needs. Conversation or discourse analysis needs Jefferson or GAT 2; thematic and content analysis usually works with simple rules. Many universities publish a recommended set.

What does (.) mean in a transcript?

In Jefferson notation, (.) is a micropause, a gap too short to time. A number in brackets, such as (0.5), is a timed pause in seconds.

Can AI transcription follow these conventions?

It produces the base text, speaker turns and timestamps. Detailed symbols for pauses, overlaps and intonation still have to be added by a person listening to the recording.

Read more

Related terms

  • Verbatim transcription

    Verbatim transcription writes down speech exactly as it was said: every um and uh, repeated words, false starts, stutters and often non-verbal sounds like [laughs]. Its opposite, clean verbatim, keeps the speaker's words and meaning but removes fillers and stumbles so the text reads smoothly.

  • Speaker diarization

    Speaker diarization is the step in speech recognition that works out who spoke when. It splits a recording into segments by voice and labels them Speaker 1, Speaker 2 and so on, so an interview or meeting transcript shows each person's turns instead of one block of text.

  • Timecode

    Timecode is a label for every frame of video, written as hours:minutes:seconds:frames — 01:02:03:12 is frame 12 of the third second. Editors use it to find the same moment in every copy of a project. Subtitle files usually count time in milliseconds instead, which is where mix-ups start.

Updated: