Free Whisper CLI transcribes 150 min of audio in 98 seconds
Insanely Fast Whisper runs locally, costs $0, and beats cloud APIs by up to 19x — no key, no billing.
Insanely Fast Whisper is an open-source CLI that transcribes audio at speeds that make paid services look absurd. With Flash Attention 2 and batching, it processes 150 minutes of audio in ~98 seconds on a modern GPU — 19x faster than vanilla Whisper, zero API costs.
It runs on NVIDIA GPUs and Apple Silicon, handles speaker diarization (who said what), translates to other languages, and outputs clean JSON with timestamps. One pipx install and you're running. Compare that to Otter.ai ($100/yr), Rev ($1.50/min), or enterprise contracts in the thousands.
The repo has crossed 11K GitHub stars with an MIT license and active development. For podcasters, journalists, researchers, or anyone doing bulk transcription — the cloud API case is basically gone.
Why it matters: local-first AI is eating SaaS pricing models. When the open-source version is free and faster, the only reason to pay is convenience — and this removes that excuse too.
Sources
Primary: the company, paper or repository
- Insanely Fast Whisper — GitHub github.com
Written by an AI pipeline from the sources above. Methodology · Report an error
Feed, daily deep-dive and bytes — readable offline, with push alerts for the topics you follow.