Available now Desktop app · runs offline

🎙️ WhisperDesk

Unlimited local audio & video transcription. Your audio never leaves your machine.

WhisperDesk does what Otter.ai charges $16.99/month for — but 100% locally, powered by whisper.cpp (OpenAI's Whisper model running natively on your CPU). Drop in any audio or video file, get back an accurate, timestamped transcript. No upload, no account, no per-minute limits.

$16.99/mo forever $39once
Screenshot on the way

WhisperDesk screenshot is being captured — the app is shipped and real.

What's in the box

Features

Drag & drop anything

mp3, wav, m4a, mp4, mov, mkv, flac, ogg, webm — video audio is extracted automatically.

100% private

Transcription runs entirely on your machine. Nothing is ever uploaded.

Unlimited minutes

No cap, no monthly quota. Transcribe 10 hours a day if you want.

Model picker

Tiny / base / small / medium, English-only or multilingual — downloaded once from Hugging Face.

Language auto-detect

Or force English. Multilingual models cover dozens of languages.

Live progress

Real-time percentage while Whisper works, with timestamped transcript view and one-click copy.

Export TXT, SRT, VTT

Ready for subtitles, notes, or your editor.

Local history

Every transcription saved locally — reload it any time.

The receipt

WhisperDesk vs Otter.ai

Otter.ai at $16.99/mo runs roughly $204/year — $408 over two years. WhisperDesk is $39, once.

WhisperDeskOtter.ai
Price$39 one-time$16.99/month, forever
Cost after 1 year$39~$204
Transcription limitUnlimited1,200 min/mo (Pro)
Max file lengthUnlimited90 min per conversation
PrivacyAudio never leaves your PCUploaded to their cloud
Works offlineYesNo
Video filesYes (audio auto-extracted)Limited
SRT/VTT subtitle exportYesPaid tiers

Pays for itself in under 3 months of Otter.ai Pro.

Setup

Three steps, no subscription

STEP 01

Buy once on Whop

One-time $39 for the packaged Windows installer — no terminal, no build step.

STEP 02

Pick a model

On first run WhisperDesk fetches whisper.cpp (~8 MB) and your chosen model (78–148+ MB) with a visible progress bar. One time only.

STEP 03

Drop a file, get a transcript

Timestamped text in minutes, exportable as TXT, SRT or VTT — all offline.

FAQ

Honest answers

Is it really free on GitHub?

Yes — MIT-licensed at github.com/bensblueprints/whisper-transcriber, and it always will be. The $39 buys the signed installer, 1-click setup and updates.

How accurate is it?

It runs OpenAI's Whisper models — the same family behind most modern transcription tools. Bigger models (small/medium) are more accurate and slower; you pick the trade-off.

Does it do live meeting transcription?

No — and Otter is genuinely better for that. WhisperDesk transcribes files (recordings, interviews, videos, podcasts). If you need a live meeting assistant with speaker chat, keep Otter.

How fast is it?

Depends on your CPU and model size. The tiny/base models typically run faster than real time on a modern machine; medium trades speed for accuracy.

Does it need the internet?

Only twice, ever: the one-time whisper.cpp binary download and your chosen model from Hugging Face. After that it's fully offline.

Deep-dive comparisons:

Own WhisperDesk forever

$39 once. Signed installer, 1-click setup, updates included. No renewal, no account with us, no meter. Or build it yourself from the MIT source — it's the same app.