
WhisperDesk turns any audio or video file into an accurate, timestamped transcript — entirely on your own machine, powered by whisper.cpp (OpenAI's Whisper running natively on your CPU). Otter.ai costs $16.99/month — over $204 a year — caps you at 1,200 minutes a month and 90 minutes per file, and uploads every recording to their cloud. WhisperDesk is $39 once: unlimited minutes, unlimited file length, total privacy.
Built for journalists protecting sources, podcasters, students, and researchers with confidential interviews.
Drag & drop anything — mp3, wav, m4a, mp4, mov, mkv, flac, ogg, webm (video audio auto-extracted via bundled ffmpeg)
Model picker: tiny / base / small / medium, English-only or multilingual, downloaded once
Language auto-detect or force English
Live progress bar while Whisper works
Timestamped transcript view with one-click copy
Export TXT, SRT, and VTT — subtitle-ready
Local history of every transcription, reload any time
Zero telemetry; fully offline after the one-time model download
Runs as a Windows desktop app (Electron).
What's included:
Windows installer download (Whisper Transcriber Setup 1.0.0.exe)
Full source access — https://github.com/bensblueprints/whisper-transcriber (MIT)
Lifetime v1 updates
Setup guide (README quick start)
Premium tiers coming later