what deposition format actually is
deposition transcripts have a fixed shape: 25 lines of testimony per page, line-numbered, page-numbered, with Q. and A. prefixes for the questioning attorney and the witness, with a caption block at the start identifying the case and parties. objections appear inline. the format is not a stylistic preference — it's how depositions get filed, cited, and impeached at trial.
search for "deposition transcript format" today and you get PDF guides explaining the format, court-website pages requiring it, and a few PDF-fillable templates. you do not get a tool that produces it from audio. that's the gap.
what we ship
a transcription mode that emits deposition-formatted output directly:
- 25-line page layout with proper line numbering down the left margin and a page footer with case number and page count.
- Q./A. speaker prefixes from speaker-diarization output. flagged for any line where the model isn't confident which voice is which — paralegal review is the second pass.
- caption block at the start of the transcript, populated from a one-time form: case name, case number, court, date, location, witness, parties, attorneys present.
- objections inline detected from "objection," "objection, form," "objection, foundation" patterns and rendered in the standard indentation. flagged for review where ambiguous.
- court-rule profiles for jurisdictions with non-default rules (a few state courts use 24- or 26-line pages, certain federal districts have different margin rules). pick the profile at export.
workflow
- drop the deposition audio file. fill the caption block (one-time per case — saved for the next deposition in the same matter).
- transcription runs. on a 60-minute deposition the first pass is ready in a few minutes.
- review in the editor. flagged speaker calls and flagged objection patterns sit at the top of a "needs eyes" panel. click any word to verify it against the audio.
- export to deposition-format PDF, .docx, or plain text. a secondary export produces a per-line CSV for use in litigation-management systems.
where this fits
this is for the working transcript that goes into a file — for trial prep, for summary judgment briefs, for impeachment binders. for the official record certified by a court reporter, the court reporter's transcript remains the official record. we don't replace court reporters. we replace the time a paralegal spends turning audio into a working draft.
what we don't claim
court-grade accuracy from any AI transcription tool. ours included. depositions have crosstalk, mumbled words, jargon-heavy testimony, and witnesses who answer before the question ends. we get to a working first pass faster; the paralegal still verifies it.
our deposition mode is english-only at launch. for non-english witness testimony, the cloud transcription mode produces a plain transcript that you can format manually using the same PDF profile.
privacy
for depositions you don't want sitting on a third-party server, run the file in private mode. everything described above runs identically; the audio just stays on your laptop.