Transcription

I turn any recording, a hearing, a deposition, an interview, a phone call, or body-cam footage, into a diarized transcript with a timestamp on every word, so you can jump straight to the moment something was said instead of reading a full recording to find it. Send me the audio or video and I get started.

What you get

• A time-coded transcript, speaker by speaker, with a timestamp attached at the word level, not just a rough time-block guess.
• A searchable table version of the same transcript loaded into your case database alongside your documents, so a question typed into the assistant can pull an answer from what was said in a recording, cited to the speaker and the timestamp, the same way it cites a document to page.
• A QA pass on the raw machine transcript before delivery. I listen against the output, correct names, places, and garbled passages, and confirm speaker labels are assigned to the right person, rather than handing over an unreviewed machine draft.

How it fits with a database build

A transcript loaded this way behaves like any other source in the database: it becomes searchable the same way your documents are, by a word, a name, a date, or what was said. At Level 2 and up, statements from a transcript can appear in the chronology and the roster of people and organizations alongside what the paper record shows, so a recorded statement and a document about the same event sit next to each other in one search.

What I need from you

The audio or video file itself, in whatever format you have it (a standard recording, a phone-call export, or body-cam footage), and, if you have one, a list of who was present, so speaker labels come back correct on the first pass rather than as "Speaker 1" and "Speaker 2."

Price

Quoted per audio hour, in writing, before work begins.

Tips on using the assistant are provided at signing. No legal advice; I transcribe and index what was said, and counsel decides what it means for the matter.