Most rejected court transcripts fail on procedure, not content. The audio was clear, the wording was correct, but a missing certificate, a typo in the case number, or an unverified speaker tag gave opposing counsel an easy procedural win. This guide is a practical reference for litigators who file transcripts as exhibits — the seven mistakes that surface most often, and the fix for each.
1. Not actually verbatim
“Verbatim” means every spoken word, including filler (“um”, “you know”), false starts, crosstalk, and side comments. Some transcription services silently smooth these out — a fatal error for cross-examination, where the cadence of an answer is the evidence. Fix: instruct the vendor in writing that the deliverable is verbatim with disfluencies preserved, and check a 5-minute sample against the audio yourself before accepting delivery.
2. Wrong or missing speaker identification
Multi-party recordings (depositions, conferences, board meetings) must label every speaker correctly. Common errors: two co-counsel labeled as the same person, a witness mis-attributed to opposing counsel, an off-record voice (court reporter, paralegal) included as a witness. Fix: at the start of the recording, each speaker should state their name, role, and relationship to the case for the record. The transcriber uses that statement as the speaker key — and so does the trier of fact.
3. Inconsistent timestamps
Timestamps locate quoted passages back in the source audio. They must be present, evenly spaced (every 30 seconds is a strong default), and based on the source file’s actual time, not the transcriber’s edit clock. Fix: require timestamps every 30 seconds and at every speaker turn, in [HH:MM:SS] format. If the audio was paused mid-recording, mark the pause explicitly.
4. Hiding inaudible passages
When the audio is unclear, the transcriber must mark the gap explicitly: [inaudible 00:14:32] or [unintelligible — overlapping speech]. Vendors who fill gaps with their best guess — without the marker — risk the entire transcript being attacked at admission. Fix: require explicit inaudible markers in the contract. A transcript with too few markers is a red flag.
5. Weak certificate of accuracy
The certificate is the procedural lynchpin of an admissible transcript. A weak certificate — missing the transcriber’s qualifications, the date, the audio file’s identifying details, or the transcriber’s signature — gives opposing counsel an easy challenge. Fix: the certificate should include (a) transcriber name and credentials, (b) statement that the transcript is true and accurate to the best of their knowledge, (c) date and physical signature, (d) audio source identification (filename + hash), (e) total duration of audio reviewed, (f) optional notarization for higher-stakes filings.
6. Court pagination errors
Federal and most state courts require specific pagination: 25 lines per page, line numbers in the left margin, double-spaced, specific font. A transcript laid out for human reading (single-spaced, no line numbers) cannot be filed without re-formatting — and re-formatting introduces drift between page references in your motion and the actual transcript. Fix: request court-formatted output (DOCX with line numbers + PDF/A) at order time, not after delivery.
7. No chain of custody
If opposing counsel asks “who had access to this file between the recording and the transcript?” — you need a documented answer. Chain of custody covers: who recorded the audio, who transmitted it to the transcriber, hash of the file at each handoff, who edited the transcript, who reviewed it, and where the master is stored. Fix: at intake, hash the audio (SHA-256), log every transfer with timestamp + hash, store originals in a write-once medium. Court & legal transcripts at Lessrec ship with the chain of custody log included.
Pre-filing checklist
Before filing the transcript as an exhibit, verify:
- [ ] Verbatim with all disfluencies preserved
- [ ] Every speaker correctly identified at first appearance and at each speaker turn
- [ ] Timestamps every 30 seconds and at each speaker turn, [HH:MM:SS] format
- [ ] Inaudible passages explicitly marked with location
- [ ] Certificate of accuracy: name, credentials, date, signature, audio identification, duration
- [ ] Court formatting: 25-line page, line numbers, double-spaced, PDF/A
- [ ] Chain of custody log attached as appendix
- [ ] File hash matches the original recording at every handoff
- [ ] Notarization (if filing requires sworn declaration)
For pricing on certified, court-formatted transcripts see transcription pricing — Lessrec includes timestamp / speaker ID / certificate by default at the legal tier.
FAQ
Is a notarized transcript admissible in US federal court?
A certificate of accuracy is the typical baseline; notarization adds extra authentication. Confirm your jurisdiction’s specifics.
How do I prove chain of custody for an audio recording?
Document who received the file, hash at intake, handoff log, audit trail of every editor.
Can opposing counsel challenge a verbatim transcript?
Yes — typically by attacking accuracy, qualifications, or chain of custody. Clean intake makes most of these challenges fail procedurally.
Have a recording? Send it.
Upload audio or video. We’ll send a transparent estimate within an hour and confirm the deadline before you pay.
Upload audio