VideoText workflow guide
The complete GoTranscript style guide for freelance transcribers. Covers clean verbatim requirements, job-specific timestamping [00:00:00], bold speaker labels, inaudible/unintelligible tags, and every formatting rule you need to pass QA and maintain high ratings.
Clean verbatim is the common case: remove speech errors, stutters, repetitions and false starts that add no information, but keep repetitions used for emphasis. Expand slang — gonna becomes going to, yeah and yep become yes — and write Okay in full, never OK or Ok. Full verbatim keeps those features as spoken. Getting this wrong means reworking the whole file, so read the job type before the first line.
A bold label, a colon and one single space — never a tab: Mark:, Speaker 1:, Interviewer:. Use real names or descriptive roles where they are known, and prefix a question mark where you are not certain who is talking (?David:). A recording with one speaker still gets a label.
Timestamping is decided per job. Where it applies, the format is a bold [00:00:00] in full hours, minutes and seconds, placed either every 2 minutes or at every speaker change — one type per file, never both. On a segment of a longer recording the time is file-relative, so minutes 20 to 30 open at [00:20:00]. The qualification test asks for no timestamps at all.
Use [inaudible 00:00:00] where speech cannot be heard and [unintelligible 00:00:00] where it is heard but cannot be understood. Overlapping speech is [crosstalk], a separate tag rather than a kind of inaudible. Do not invent notation the guidelines do not list.
Clean verbatim strips fillers, stutters and uninformative false starts and expands slang into standard English. "Oh" stays in at either verbatim level. Spoken contractions are kept as spoken, and a speaker's grammar is never corrected — though a misspoken word is still spelled correctly.
Spell out zero through nine and use numerals from 10 up, with money, years and percentages taking numerals at any size. This one is worth checking deliberately: once the habit is wrong it repeats through the whole file.
Exclamation marks are not used, and the [sic] tag has not been part of the guidelines since 2019. A double dash (--) closes a false start or an incomplete sentence, while a single dash marks an interruption.
Keep a paragraph under roughly 500 symbols — about 100 words — and break on a topic shift inside a long turn rather than at a fixed interval.
Reduce revision risk before submitting marketplace jobs. A pre-delivery formatting check catches the label inconsistency, timestamp format error, or verbatim level mismatch that triggers a QA rejection and unpaid revision.
Create a consistent formatting baseline across a team of transcriptionists working on the same client account — so reviewers spend time on content accuracy, not fixing label formats.
Turn raw AI-generated transcripts into readable interview documents: speaker structure, consistent punctuation, accurate paragraph breaks, and timestamps that let readers verify context in the source recording.
Page 1: "JOHN SMITH: Thanks for joining us." Page 3: "John: So as I was saying..." Page 7: "Speaker 1: Right, exactly." — All three refer to the same speaker. Label drift of this kind fails QA at every major transcript marketplace.
[00:05:12] on page 2, (00:05:12) on page 4, [5:12] on page 6, and 00:05:12 on page 8 — all in the same document. Most style guides require a single format, and inconsistency triggers automatic rejection on automated QA systems.
"[inaudible]", "[INAUDIBLE]", "[unclear]", and "[crosstalk]" appearing in the same file for the same type of audio problem. Rev requires "[inaudible]" in lowercase brackets; GoTranscript requires a full timestamp inside the tag and keeps [crosstalk] separate. Neither accepts a mix of styles.
Pages 1–4 are clean verbatim (fillers removed). Pages 5–8 switch to full verbatim (fillers preserved). This happens when an ASR tool applies different post-processing across transcript segments — and it fails QA because the verbatim level is supposed to be consistent throughout.
Speaker labels, timestamps and every time-stamped tag must be bold. Most platforms treat transcript formatting as cosmetic, so draft in an editor that preserves bold and confirm it survived the export.
The split between [inaudible] for speech that cannot be heard and [unintelligible] for speech heard but not understood is unusual — most services use a single unclear-audio tag, and transcribers arriving from them tend to collapse the two.
Periodic stamps and per-turn stamps are house defaults elsewhere. GoTranscript picks between them job by job, so the same transcriber can owe stamps on one file and none on the next.
Notes go in square brackets, lowercase, present tense, two words at most: [crosstalk], [silence], [background noise], [laughs]. Only time-stamped marks such as [pause 00:00:00] carry a time.
It depends on the job. When timestamping is required, use bold [00:00:00] (full hours:minutes:seconds) either every 2 minutes or at every speaker change — check the job instructions. The qualification test is clean verbatim with no timestamping. Missing timestamps on a job that requires them will cause QA failures.
Most jobs use clean verbatim: remove filler words (um, uh, you know, like), false starts, and repetitions (keep emphasis repetitions). Expand slang (gonna → going to). Keep "Oh" expressions and spoken contractions. Some jobs use full verbatim — check the job type before you start.
Speaker labels are bold, followed by a colon and a single space: Speaker 1:, David:, Interviewer:. Use names or descriptive roles when known. If you cannot identify who is speaking, prefix a question mark: ?David: or ?Speaker 3:. Always label even single-speaker files.
Use [inaudible 00:00:00] when speech cannot be heard (noise/recording) and [unintelligible 00:00:00] when speech is heard but not understood (accent/manner). Both use full HH:MM:SS timestamps and must be bold. Do not invent other markings.
GoTranscript editors rate accuracy on a 1–5 scale (96–100% = score 5). There is no published "minimum accuracy percentage" on their guidelines page. Consistent formatting compliance and strong ratings determine continued access to jobs.