Clean Verbatim vs. True Verbatim: What's the Difference — and Which Does a Court Transcript Need?
The two general-transcription styles
Outside the courtroom — in business, research, media, and academic transcription — there are two standard products, and the whole industry defines them the same way:
True verbatim (a.k.a. full or strict verbatim)
Everything the microphone caught, on the page. Filler words ("uh," "um," "you know," "like"), false starts, stutters, repeated words, interjections ("uh-huh," "yeah"), and non-verbal events ([laughs], [coughs], [pause]) are all transcribed. Nothing is smoothed. It is the truest representation of how something was said, which is exactly why it's used where manner matters — recorded statements, research on speech, and testimony where a hesitation may be evidence.
Clean verbatim (a.k.a. intelligent verbatim, or a "clean read")
The same words, cleaned for readability. Fillers, false starts, stutters, throat-clearing, and unintentional repetition come out; the speaker's actual words, order, and meaning stay in. The one hard rule is what separates it from summarizing: clean verbatim never paraphrases. It removes noise, not substance — you don't rewrite a sentence, fix its grammar, or condense a rambling answer. You take out the "um" and the accidental "the, the," and leave everything the person actually said.
| What happens to… | True verbatim | Clean / intelligent verbatim |
|---|---|---|
| Filler sounds (uh, um, you know) | Kept | Removed |
| False starts ("I—I turned left") | Kept | Removed |
| Repeated words ("the, the car") | Kept | Removed |
| Stutters | Kept | Removed |
| Non-verbals ([laughs], [pause]) | Kept / noted | Removed |
| The actual words, in order | Kept | Kept |
| Grammar & sentence structure | As spoken | As spoken (never rewritten) |
| Meaning | Kept | Kept |
Why a court transcript is a third thing
Here's where the general definitions mislead people. Search "clean vs. true verbatim" and you'll be told to pick one. For a certified court or deposition transcript, the honest answer is you pick neither — the filed record is its own point on the spectrum, and it matters that you know why.
Full clean verbatim removes too much for the legal record. A false start or a repeated word can carry legal meaning — "No -- I mean, yes" is the classic example, but so is a witness who starts to name one person and switches to another. The official record has to preserve the words actually spoken, so a certified transcript keeps false starts and repetitions. It just writes them the professional way, with the interruption dash rather than a hyphen weld — a convention we counted across hundreds of certified pages in a separate guide on repeated words and stutters (652 dash-marked repeats in Supreme Court transcripts and 85 in a federal Michigan record, with zero hyphen-welds).
Strict true verbatim is usually more than the filed record carries. The clearest published rule on the point — the Judicial Conference's transcript format policy for the federal courts — licenses dropping hesitation sounds:
"In the interest of readability, however, false starts, stutters, uhms and ahs, and other verbal tics are not normally included in transcripts; but such verbalizations must be transcribed whenever their exclusion could change a statement's meaning." Guide to Judiciary Policy, Vol. 6, Ch. 18, § 18.10.1.b
And certified practice matches: in a survey of 20+ certified transcripts from the public record — Supreme Court arguments, federal trials, Michigan hearings and depositions — the overwhelming majority contain zero standalone "uh"/"um," even in hours of live testimony. (The full count, the exceptions, and the one thing that always stays — "uh-huh" and "uh-uh," which are answers, not fillers — are in our guide to filler words.) So the filed record drops hesitation sounds like clean verbatim, but keeps every real word like true verbatim.
| What happens to… | True verbatim | The certified court record | Clean / intelligent verbatim |
|---|---|---|---|
| Filler sounds (uh, um) | Kept | Dropped for readability | Removed |
| False starts & repetitions | Kept | Kept, with the " -- " dash | Removed |
| Answer forms (uh-huh, uh-uh) | Kept | Kept — they're answers | Kept |
| Every actual word, in order | Kept | Kept | Kept — never paraphrased |
| Right for the official filed record? | Valid, but stricter than most | Yes — the dominant convention | No — it removes words that are part of the record |
The same answer, three ways
Take one line of live deposition testimony. Here is what the witness actually said, then the same answer in each style:
Every version keeps "Well" (it's a real word) and none of them rewrites the sentence. The difference is only how much noise comes off: true verbatim keeps the "um," the "uh," and the two false starts; the court record drops the two hesitation sounds but keeps the false starts as "I -- I" and "the -- the"; clean verbatim takes all of it out for a smooth read. Notice what clean verbatim does not do — it doesn't shorten "turned left on Main Street" or fix anything. Remove more than the noise and you've stopped transcribing and started editing.
Which one do you need?
- The official court or deposition transcript you certify: the middle column. Keep every word, write false starts and repeats with the dash, keep "uh-huh"/"uh-uh" answers, and drop standalone hesitation sounds — dropping is the dominant certified convention, and keeping them is a permissible minority practice if that's your standard.
- True verbatim: when a client orders it, and when manner of speech is itself the evidence — recorded statements, examinations under oath, interpreter proceedings, or anything where a hesitation may matter. It's also the safest base for a certifying reporter: you can always drop a sound on review, but you can't recover a word you never captured.
- Clean verbatim: for readable, non-record work — a witness-prep read, an internal summary copy, business or research transcription, media subtitles. Excellent for readability; wrong for the filed legal record, because it removes words that are part of that record.
- Whatever you choose, be consistent. A transcript that cleans one speaker and not another reads as editorializing — the one thing a reporter must never appear to do. Pick the rule, apply it to every speaker on every page, and let the certificate mean what it says.
Sources
- Guide to Judiciary Policy, Vol. 6, Ch. 18 — Transcript Format (§ 18.10.1.b, "Editing of Speech" — the readability license and its meaning-based limit)
- Stenobox, Do court reporters include "uh" and "um"? — the counted certified-transcript survey and the answer-forms rule
- Stenobox, How court reporters write stutters and repeated words — why the record keeps them, and the dash convention (counted)
- The definitions of "true / full verbatim" and "clean / intelligent verbatim" are standard usage across the transcription industry (transcription providers, training bodies, and legal-transcription references describe them the same way).
Common questions
What is the difference between clean verbatim and true verbatim?
True verbatim captures everything — fillers, false starts, stutters, repetitions, non-verbal sounds. Clean verbatim keeps the exact words and meaning but removes that noise, without ever paraphrasing. Clean edits out sounds and slips; it does not touch substance.
Is intelligent verbatim the same as clean verbatim?
Yes — "clean verbatim," "intelligent verbatim," and "clean read" all name the same style: readable, every real word kept, fillers and false starts removed, no paraphrasing.
Which verbatim style does a court transcript use?
Neither of the general styles exactly. The certified record keeps every word — false starts and repetitions included, written with the dash — but drops standalone hesitation sounds. That's the dominant convention; keeping every filler is a permissible minority practice, and some clients order true verbatim outright.
Does clean verbatim remove false starts and repetitions?
Yes — and that's precisely why full clean verbatim isn't used for the official record. A false start or repeated word can change what testimony means, so the filed transcript keeps them; clean verbatim takes them out.
When should you use true verbatim?
When how something was said matters as much as what was said — recorded statements, sworn examinations, interpreter proceedings — and as the safe starting point for a certifying reporter, since a sound can be dropped later but a missed word can't be recovered.
This guide reports a published court policy and standard industry conventions for general information. Whether a particular record must be true or clean verbatim can depend on the court, the client's order, and state rules — confirm the requirement for your work before relying on it. This is not legal advice.
Related guides: Do court reporters include "uh" and "um"? · How to write stutters and repeated words · All Stenobox guides →