Transcribe Audio Evidence for Discovery With a Prep Checklist

October 26, 2026 · 6 min read

To transcribe audio evidence for discovery, preserve the original recording untouched and work from a copy. Gather the context a transcriber needs, such as speaker names, key terms, and the portions that matter. Then decide on transcript conventions before anyone starts typing. This guide is for small law firms, solo practitioners, and other US professionals who need a clear, reviewable written record of a recording that may be produced, quoted, or challenged.

How to prepare audio files before transcription starts

Start with the file itself. Save the original recording exactly as you received it, and never edit, trim, or "clean up" that version. If you want to boost volume or reduce noise to help with listening, do it on a separate working copy and label it clearly.

Next, document what you have. Note the file name, format, length, the source, and the date you received it. Your attorney or your firm's evidence-handling procedures should govern how you store and log these files. The transcription step should fit into those procedures, not replace them.

Finally, think about scope. Do you need the entire recording transcribed, or specific segments? If only certain sections are relevant, write down the start and end times. That said, many legal teams prefer a full transcript so nothing is taken out of context.

What to check before you send recordings for transcription

A transcriber can only be as accurate as the information they start with. Before sending anything out, run through this checklist:

• Original preserved: The untouched source file is stored separately from any working copy.
• File plays fully: You have listened to the start, middle, and end to confirm the file isn't corrupted or truncated.
• Speaker list: You have the names and roles of everyone likely to speak, with spellings confirmed.
• Glossary: You have listed case-specific terms, such as company names, product names, addresses, medical terms, or slang.
• Known problem spots: You have noted timestamps where audio drops, people talk over each other, or background noise is heavy.
• Format requirements: You know whether your team wants verbatim or clean verbatim, timestamps, and how speakers should be labeled.
• Confidentiality instructions: You have written down any handling requirements your firm or client expects.
• Deadline: You know the real date the transcript is needed, with room left for your own review.

How to choose transcript conventions for evidence

Transcripts used in discovery are read closely, sometimes by opposing counsel looking for weaknesses. Consistent conventions make your transcript easier to defend and easier to cite. Agree on them in writing before work begins.

The core principle is simple: a transcript should never fill gaps with guesses. If a word can't be heard with confidence, it should be marked, not invented.

Common mistakes and how to avoid them

Editing the only copy of the recording

Noise reduction or trimming applied to the original can raise questions about integrity. Always keep the source file unaltered and work only from labeled copies.

Relying on an automatic draft as the final record

Speech-recognition drafts can be a starting point, but they struggle with crosstalk, accents, phone audio, and proper names. They can also produce fluent-sounding text that the speaker never said. For evidence, every passage needs to be checked against the audio by a person.

Skipping the speaker list and glossary

Without context, a transcriber has to guess at names and terms. That leads to inconsistent spellings across the transcript. A short reference document prevents most of these errors.

Mislabeling speakers

Assigning a statement to the wrong person can change what a transcript appears to show. If identity is uncertain, use neutral labels and let your legal team make the attribution.

Leaving no time for review

Even a carefully edited transcript should be spot-checked by someone who knows the case. Build that review into your timeline instead of reading the transcript for the first time the night before it's due.

What you can do yourself vs. when to use a professional transcription service

Some recordings are reasonable to handle in-house. A short, clear call between two people, needed for internal reference, may not justify outside help. If someone on your team has the time and a careful ear, transcribing it yourself can be practical.

Outside help makes more sense when recordings are long, when several people talk at once, when audio quality is poor, or when the transcript may be produced or quoted. It also helps when your team's hours are better spent on legal work than on replaying the same ten seconds of audio.

For example, imagine a small firm that receives a long recorded meeting with several participants, some on speakerphone, as part of a document production. A paralegal could start transcribing it, but the crosstalk and unfamiliar names would make progress slow and the result hard to verify. In that situation, the firm might send a working copy, a speaker list, and a glossary to a human-edited transcription service. It would then keep its own staff focused on reviewing the finished transcript against the case issues.

Whichever route you choose, someone on your side should still verify key passages against the audio. A transcription service supports your review; it does not replace your judgment about what the evidence means.

How to send recordings and instructions to a transcription provider

Put everything in writing. Include the file details, the scope, the conventions you chose, the speaker list, the glossary, and your deadline. Written instructions create a record of what was requested and reduce back-and-forth.

Ask how unclear passages will be marked, and how questions will be raised during the work. Ask too what the final deliverable will look like. Getting these answers before work starts helps you avoid surprises when the transcript arrives.

At LessRec, we prepare human-edited English transcripts of legal recordings for US clients. Every order starts with an individual quote. You share your recording and requirements, we review the material, and a manager confirms the quote in writing. Once you place an order, an editor listens to the full recording and checks the entire transcript against it, flagging unclear audio rather than guessing. Request an individual quote and tell us what your discovery materials need.

Related articles

• How to get accurate transcripts of dictated doctor's notes
• Deposition transcript checklist before you quote testimony
• Telehealth transcription tips for accurate session records

FAQ

Should I edit or clean up the original audio before transcription?

No. Keep the original recording exactly as you received it, and apply any volume boost or noise reduction only to a separate, clearly labeled working copy.

Should a discovery transcript be full verbatim or clean verbatim?

Many legal teams choose full verbatim because hesitations, false starts, and self-corrections can matter to meaning. Whichever style you pick, agree on it in writing before work begins.

How should unclear words be marked in an evidence transcript?

Use tags such as [inaudible] or [unclear] with a timestamp. A transcript should show where the record is uncertain instead of filling gaps with guesses.

Can I use an automatic speech-recognition draft as the final transcript?

It can be a starting point, but these drafts struggle with crosstalk, accents, phone audio, and proper names, and may produce fluent text no one said. For evidence, a person should check every passage against the audio.