Transcribe a Zoom recording without a monthly subscription

May 5, 2026 · 4 min read

Zoom AI Companion is bundled with paid Zoom plans (Pro and above) and gives you a transcript automatically. Otter integrates with Zoom and gives you live captions plus a transcript. Both work — but they assume you record meetings constantly. If you record one Zoom every two weeks, you're paying $20-30/mo for an empty seat.

Here's the cheap path.

Cost comparison for one 1-hour Zoom

The 4-step workflow

1. Record the Zoom locally

In Zoom: Settings → Recording → Record to this computer. The host gets the option to start recording mid-call. After the meeting, Zoom drops three files into your local Recordings folder: video_*.mp4, audio_only.m4a, and a chat.txt.

You only need the audio_only.m4a file. It's about 30-50 MB per hour at default Zoom audio quality (44 kHz mono).

2. (Optional) compress further if file is huge

Zoom recordings are sometimes saved at high bitrate. To shrink:

ffmpeg -i audio_only.m4a -b:a 64k -ac 1 zoom-meeting.mp3

1-hour Zoom audio drops to ~28 MB. No accuracy loss for Whisper.

3. Send it for a human-reviewed transcript

Lessrec provides human-reviewed transcription: an editor listens to the recording and checks the full text. Send your project requirements for an individual written quote.

4. Wait 1-2 minutes, download

Whisper large-v3 INT8 transcribes 1 hour of clean Zoom audio in ~50-90 seconds. You get .txt, .docx, and .srt. Open the .docx in Google Docs or Word, drop a few section headers, and you have a meeting note.

What about live captions during the call?

If you need real-time captions while the meeting is happening — LessRec doesn't do that. Use Zoom's built-in live transcription (free for hosts on paid plans) or Otter's live caption feature.

If you only need the transcript after the meeting (the more common case), the upload-and-wait flow is the right tool.

Speaker labels (who said what)

Zoom AI Companion attributes lines to participant names because it knows who's on the call. Tools that only see the audio file merge speakers into one stream; at LessRec, the editor labels speakers while checking the recording.

Lessrec provides human-reviewed transcription: an editor listens to the recording and checks the full text. Send your project requirements for an individual written quote.

Privacy: where does the audio go?

Your file lives on a Hetzner server in Helsinki (EU GDPR jurisdiction) for 7 days, then deletes automatically. We don't train on your data, ever. We don't sell it, don't share it, don't index it.

If you're transcribing sensitive meetings (HR conversations, legal calls, founder discussions) and need stricter compliance — in Telegram @audioburobot for an EU-only enterprise tier with shorter retention.

When Zoom AI Companion is better

• You record meetings 5+ times per week
• You need automatic speaker labels
• You want the transcript searchable inside Zoom Cloud directly
• Your team is already on Zoom Pro+ for other features

When LessRec is better

• You record Zooms occasionally, not weekly
• You want to pay per-meeting instead of per-month
• You want one tool that also works for non-Zoom audio (podcasts, interviews, voice memos)
• You don't want a meeting bot joining your calls

FAQ

Can I just upload the .mp4 video?

Yes, video files work, but the audio-only .m4a is smaller and faster to send.

What if my Zoom recording is on Zoom Cloud, not local?

Open Zoom web, go to Recordings, download the audio file, then send it to LessRec.

What if the meeting was in Spanish/Russian/Chinese/etc?

Price and delivery are agreed in a written quote after we review your materials and requirements.

Multiple speakers all in one audio track — does Whisper handle it?

Raw Whisper output has no speaker labels; all voices merge into one transcript. A human-reviewed transcript adds checked labels. For a 2-3 person Zoom meeting it's usually obvious from context who's speaking when.