How to Convert a Zoom Transcript to a Word Document

You need a Word copy of a Zoom transcript — to send a client a written record of the session, paste key lines into a report, or just have something you can search and edit without opening a video. You download the transcript from Zoom's web portal, open the file, and find a wall of sequence numbers and timestamps with your sentences chopped into fragments between them. That's not a document. It's a captioning file, and turning it into something you'd actually hand someone takes real cleanup.

What's actually happening

Zoom's cloud recording can generate an audio transcript for a meeting — a setting you (or your account admin) turn on under the account, group, or personal Recording settings, in the Advanced cloud recording section as "Create audio transcript." Once it's on, Zoom transcribes the spoken audio for that meeting and, per Zoom's own documentation, "appears as a separate VTT file in the list of recorded meetings." Two things about that are easy to miss: the setting only applies going forward — it can't be turned on retroactively for a meeting you already recorded — and it requires cloud recording to be enabled for that meeting in the first place, since the transcript is generated from the cloud copy.

The format itself is WebVTT, a W3C standard built to drive closed captions in a video player, not to be read on its own. A real Zoom transcript looks like this:

WEBVTT

1
00:00:03.120 --> 00:00:06.480
Priya Anand: The main thing I want to flag before

2
00:00:06.480 --> 00:00:09.760
Priya Anand: we get into the budget numbers is

3
00:00:09.760 --> 00:00:11.200
David Klein: Sure, go ahead.

Every three-to-five seconds of speech is its own numbered cue with its own timestamp range. A single sentence routinely spans two or three cues. Word will open a .vtt fine — it's plain text — but what you'll see is exactly this structure, not a readable transcript.

How to convert it by hand

If you're doing this manually, the steps are:

1. Download the .vtt from Zoom. In the web portal, go to Recordings and Transcripts, open the meeting, and find Audio Transcript in the list of files. Click Download and it saves to your computer as a .vtt file.
2. Open it as plain text. Most word processors, including Word, will open a .vtt directly, or you can rename the extension to .txt first.
3. Strip the cue numbers and timestamp lines. This is the tedious part — either delete every sequence-number and --> timestamp line by hand, or run a find-and-replace pass to remove them. For a 10-minute call that's a few dozen lines; for a full hour it's several hundred.
4. Merge the fragments and clean up speaker labels. Once the cue lines are gone, you're left with one line per fragment, still broken mid-sentence. Combine consecutive lines from the same speaker into paragraphs, and reformat the speaker name however you want it to read — bold, a colon, a new paragraph, whatever your document style calls for.

None of this is hard, but it doesn't scale. One transcript is an annoying ten minutes. A full week of client sessions is an afternoon you don't have.

How RecordFlow fits

If your RecordFlow archive is on OneDrive or Dropbox, every synced recording gets a real .docx transcript, rendered directly — no Google Doc step, no download-and-convert. It's built the same way you'd clean one up by hand: a heading, the meeting date, a speakers list, then one paragraph per cue with a rounded [HH:MM:SS] timestamp and the speaker's name bolded, cue numbers and fragment breaks already gone. It lands next to the video and the raw .vtt every time a recording syncs, and the transfer goes straight from Zoom to your own OneDrive or Dropbox — RecordFlow doesn't keep a copy anywhere in between. If your archive is on Google Drive instead, you get the same clean transcript as a native Google Doc, which downloads to .docx in two clicks from Drive — see the companion guide on getting Zoom transcripts into Google Doc and AI tools for that path.