Transcript modes

Clean verbatim vs full verbatim transcription

Two transcripts of the same recording can both be accurate and still read very differently. The difference is the mode. Here is what each one keeps, the same interview done both ways, and how to pick the right one for research, legal and meeting work.

The short answer Clean verbatim (also called intelligent verbatim) keeps every word that carries meaning and takes out the clutter of speech: fillers like um and uh, stutters and repeated false starts. Full verbatim (also called strict verbatim) keeps everything exactly as it was spoken. Choose full verbatim when how something was said matters as much as what was said. For nearly everything else, clean verbatim is the better read.

What clean verbatim is

Clean verbatim records what the speaker said, made easier to read. It is the default in most professional transcription, and the names clean verbatim and intelligent verbatim describe the same thing.

A clean verbatim transcript takes out the parts of speech that carry no substance:

  • fillers such as um, uh, er and you know, where they add nothing;
  • repeated false starts, so "I just want to, to understand" becomes "I just want to understand";
  • stutters and incidental repetitions.

Just as important is what it keeps. Meaningful pauses, laughter, interruptions and colloquial phrasing stay in. So do grammar as it was spoken, dialect and slang, and a word from another language when a speaker reaches for one, which in South African speech happens constantly. Clean verbatim removes clutter. It does not rewrite the speaker.

What full verbatim is

Full verbatim, also called strict verbatim or true verbatim, records speech as it was spoken. Nothing is smoothed. A full verbatim transcript keeps:

  • every filler, repetition, false start and stutter;
  • significant pauses, often with their length;
  • overlapping speech, where two people talk at once;
  • short bracketed markers for non-verbal events that affect the meaning, such as [laughs] or [long pause].

It reads less smoothly, by design. The hesitations are the data.

The same interview, both ways

Here is one short exchange from a demonstration research interview, transcribed in each mode. The highlighted words are the ones full verbatim keeps and clean verbatim takes out.

Clean verbatim Researcher · 00:00:41

Where did you find support when the workload felt demanding?

Maya · 00:00:49

A classmate and I met at the library in Mowbray on Saturday mornings. That made a big difference because I could ask questions before I felt behind.

Full verbatim Researcher · 00:00:41

Okay. And where, where did you find support when the workload felt, felt demanding?

Maya · 00:00:49

A classmate and I, uh, met at the library in Mowbray on Saturday mornings. That made a big difference because I could ask questions before I, I felt behind.

Demonstration extract. The names and the interview are fictional.

Both versions are accurate. Both have the same words in the same order wherever the words carry meaning. The full verbatim version simply keeps a record of the delivery as well: the researcher's restart, Maya's small hesitation. Whether that record matters depends entirely on what you are going to do with the transcript.

What each mode keeps

In the recordingClean verbatimFull verbatim
Fillers: um, uh, erRemoved where they add nothingKept
False starts and repetitionsRemovedKept
StuttersRemovedKept
Significant pausesKept where meaningfulKept, often timed
Laughter, interruptionsKept where meaningfulKept
Overlapping speechKept where it mattersKept and marked
Grammar, dialect and slangKept as spokenKept as spoken
Words from another languageKept as saidKept as said
A word that will not come clearMarked with its timestampMarked with its timestamp

The last row is worth noticing. In both modes, a passage that cannot be heard should be marked, for example [inaudible 00:14:32], and a probable word as [unclear: affidavit? 00:14:32]. Neither mode is a licence to guess.

Which mode your work needs

The question to ask is not which mode is more accurate. It is whether the way something was said is part of what you are studying, relying on or quoting.

Qualitative research

For thematic analysis, which most postgraduate interview studies use, clean verbatim is usually right: you are coding what participants said, and the fillers get in the way. Discourse analysis and conversation analysis are different. Hesitations, overlaps, self-corrections and pauses are the material those methods analyse, so they need full verbatim, sometimes with even finer timing than a standard transcript carries. Narrative and interpretative approaches sit in between.

Before you choose, check what your methodology, your supervisor or your ethics protocol already specifies. A two line question now is cheaper than a second transcript later. More on qualitative research transcription.

Legal work

Where a transcript may be relied on as a record of exactly what a person said, such as a statement, a hearing or a disciplinary enquiry, full verbatim is the safer choice. Hesitation, a corrected answer or a repeated phrase can be part of the point, and a clean transcript would have removed it. For an attorney's internal record of a consultation, clean verbatim is usually enough. If the transcript is for a particular forum, check whether it has its own format or conventions first. More on legal transcription.

Meetings, interviews for publication and focus groups

Board and committee meetings, media interviews you intend to quote, and most market research focus groups are almost always better in clean verbatim. The readers were often not in the room, and they want the substance. If a meeting record is going to be read widely, an edited transcript may be better still. More on meeting and minutes transcription.

Edited and research interview transcripts

Two further modes are worth knowing about.

An edited transcript goes further than clean verbatim. Grammar and presentation are improved so the record reads well, but it stays traceable to the recording, and facts, commitments and each speaker's position are never changed. Because it is a step away from the exact words, it should be labelled as an edited transcript, so nobody mistakes it for verbatim.

A research interview transcript is a deliberately conservative approach for qualitative work: participant language and meaningful interaction markers are preserved, and names are used only where they are authorised or unambiguously confirmed. Otherwise neutral labels or agreed pseudonyms stand in.

Does full verbatim cost more?

Usually, yes, because it takes longer. Every repetition and false start has to be typed, every pause judged, and the whole thing checked against the audio at a finer grain. At Transcribe Refinery full verbatim adds 25 percent to whichever rate applies:

60 recorded minutes, Standard turnaroundRate per minuteTotal
Clean verbatimR14.50R870.00
Full verbatimR18.13R1,087.80

All prices are inclusive, and the full rules, including what more speakers, timestamps and difficult audio add, are on the pricing page, where you can work out an estimate for your own recording.

One practical consequence follows from the cost difference. A full verbatim transcript can always be cleaned later, but a clean transcript cannot be turned back into full verbatim without listening to the whole recording again. If you are genuinely unsure which your analysis will need, say so before the work starts and we will talk it through.

How to ask for the right mode

Put the mode in your brief, in one line, along with anything that goes with it: whether you want pauses timed, how speakers should be labelled, and whether timestamps should appear at intervals or at each change of speaker. If you are briefing a transcriber for the first time, our free field guide, Record It Right, has a ten line brief you can fill in, plus a checklist for recording interviews that transcribe cleanly in the first place.

Common questions

Is intelligent verbatim the same as clean verbatim?

Yes. Intelligent verbatim and clean verbatim are two names for the same mode: the speaker's words and meaning kept intact, with repeated false starts and incidental fillers taken out. Full verbatim and strict verbatim are likewise two names for one mode.

Does clean verbatim correct the speaker's grammar?

No. Clean verbatim removes clutter, not the way a person speaks. Grammar, dialect, slang and words from another language stay as they were said. A transcript that tidies grammar is an edited transcript, and it should be labelled as one.

Is full verbatim more accurate than clean verbatim?

No. Both should be equally accurate about what was said. Full verbatim records more detail about how it was said: the hesitations, repetitions and false starts. Accuracy is about getting the words right; the mode is about how much of the delivery the page keeps.

Can a clean verbatim transcript be turned into full verbatim later?

Not without listening to the whole recording again, because the detail full verbatim needs was never written down. Going the other way is easier: a full verbatim transcript can be cleaned. If you are unsure which your analysis will need, ask before the work starts.

How much more does full verbatim cost?

At Transcribe Refinery, full verbatim adds 25 percent to whichever rate applies, because it takes longer to transcribe and check. Sixty minutes on Standard comes to R870.00 in clean verbatim and R1,087.80 in full verbatim, all inclusive.

Not sure which you need?

Tell us about the recording and what the transcript is for. We will recommend a mode and reply with a written quote, usually within one working day. Nothing is sent, and nothing is charged, until you accept.

Get a written quoteWork out a price

Published 23 October 2026 by Shireen Bruinders, founder of Transcribe Refinery, updated 27 October 2026. Prices are as at publication; the pricing page always has the current rates.