Zoom Transcript Downloader

Use the zoom CLI tool to download transcripts from Zoom meeting recordings.

Important: Avoid Interactive Mode

The -s (search) flag triggers interactive prompts which don't work well in automated contexts. Always use direct meeting ID download when possible.

Usage

Direct Download (Preferred)

# Download transcript by meeting ID to stdout
zoom 123456789

# Save transcript to a specific file
zoom 123456789 -o transcript.vtt

# Save to a specific directory
zoom 123456789 -o meetings/call-transcript.vtt

Finding Meeting IDs

If you need to find meeting IDs, run the search in a way that shows the list but exits:

# List recent meetings (will show IDs in output)
zoom -s "" 2>&1 | head -20

Then use the meeting ID directly with zoom <meeting_id>.

Arguments

Argument	Short	Description
meeting_id		Meeting ID to download (positional, required for non-interactive use)
--output	-o	Output file path for the transcript
--search	-s	Search filter for topic (triggers interactive mode - avoid in automation)
--days	-d	Number of days to look back (default: 45)

Output

Returns VTT (WebVTT) transcript files containing timestamped speaker labels and text.

Example VTT format:

WEBVTT

00:00:05.000 --> 00:00:10.000
Speaker 1: Hello everyone, welcome to the meeting.

00:00:10.500 --> 00:00:15.000
Speaker 2: Thanks for joining today.

Requirements

Environment variables for Server-to-Server OAuth:

• ZOOM_CLIENT_ID
• ZOOM_CLIENT_SECRET
• ZOOM_ACCOUNT_ID

The hamel package must be installed: pip install hamel

Examples

Download a specific meeting transcript:

zoom 85678123456 -o meeting-notes.vtt

Download and immediately read the content:

zoom 85678123456 > transcript.vtt && cat transcript.vtt

Process transcript with AI:

zoom 85678123456 -o /tmp/meeting.vtt
ai-gem "Summarize this meeting transcript" /tmp/meeting.vtt