How to Transcribe a Zoom Recording

Local recording or cloud recording?

Zoom offers two ways to record, and they decide where your file ends up.

Recording to your computer

Available on free and paid accounts in the Zoom desktop app (not on phones or tablets). After the meeting ends, Zoom converts the recording and saves it to a folder on your computer, by default a Zoom folder in your Documents.

Cloud recording

Available on paid accounts. The recording is stored in your Zoom web portal under Recordings, from where you can play, share or download it. Some paid plans can also create an audio transcript of cloud recordings automatically.

If you aren't the host, you need the host's permission to record locally; account admins can also control who may record. Recording settings, including saving closed captions as a VTT file with computer recordings, are in the Zoom web portal's recording settings.

Which file to transcribe

A local recording folder typically holds an MP4 with the video and audio, and, depending on your settings, an audio-only M4A. Either works for transcription. The audio-only file is far smaller, so it uploads in seconds. If you enabled the option to record a separate audio file for each participant, you'll also find one file per person, which is handy for interviews: transcribe each speaker's file to know exactly who said what.

From the cloud, download the MP4 or the audio file from the Recordings page. Very long meetings produce large video files; if a file is unwieldy, extracting the audio first gives you the same transcript from a much smaller upload.

Zoom's own transcripts versus a separate tool

Zoom's cloud recording transcripts, where your plan includes them, are convenient: they appear next to the recording. A separate tool makes sense when:

• You recorded locally, on a free account, or on a plan without transcripts.
• The meeting was in a language Zoom's transcript doesn't support, or you need a translation.
• You need subtitle files (SRT or VTT) to publish the recording with captions.
• You want the transcript in a particular format, such as plain text with timestamps for research coding.

Step by step with mydubly

1. Find the recording: the MP4 or audio-only M4A in your Zoom folder, or a download from the web portal.
2. Open video to text (or audio to text for the M4A) and add the file. A video stays on your device; the browser extracts the audio and sends only that.
3. Choose the language spoken, or another language to receive a translated transcript.
4. Download the plain transcript, the timestamped transcript and SRT or VTT subtitles.

Transcripts cost 1 credit per minute (a tenth of a cent), with 100 free credits when you sign in with Google, and recordings up to 2 hours. A 45-minute meeting costs 45 credits.

Consent and privacy

Zoom shows participants that a meeting is being recorded, but consent to recording isn't the same as consent to having the recording transcribed by a third party, especially for research or client calls. Tell participants how the recording will be processed and stored. With mydubly, the audio is sent over HTTPS and deleted with the results within 30 minutes of delivery; cloud speech processing privacy lists questions to ask any service.

Limits to plan around

• mydubly transcripts don't label speakers. For interviews, Zoom's separate per-participant audio files are the cleanest workaround; otherwise, add names while proofreading. What speaker diarization is explains why automatic labels are hard.
• Crosstalk, poor microphones and echo reduce accuracy. Recording a remote interview covers settings that help next time.
• Recordings longer than 2 hours need splitting first.
• Live transcription during the meeting is a different job; mydubly works on recordings after the call.

Translating a Zoom recording for people who weren't there

International teams and classes often have members who follow more easily in their own language. From the same Zoom file, run a translation: choose German, Japanese, Hindi or another of 21 languages, and you get a translated transcript and subtitles. If the recording is worth watching rather than reading, a dubbed version replaces the voices with translated ones in each speaker's own voice, for 50 credits per minute; the video translator covers both. Translating videos with multiple speakers explains how a dub handles several voices.

After the transcript

Turn the transcript into minutes, action items or an article. Meeting recording to notes covers summarising, and the meeting transcription page covers the recurring workflow for teams. Proofread names and numbers before sharing; how to proofread an AI transcript gives a fast order.