Audio and podcasting decision
MacWhisper
A capable developer can recreate the core local speech-to-text workflow in about a week using existing open-source Whisper/STT projects; only product polish and commercial packaging are left as paid conveniences.
Visit website↗Open-source builds that already do this
Every project below is open source and already does this job today. Fork one, self-host it, or take the parts you need — the build prompt further down assumes an empty file, and this is the shortcut past that. Licences differ; check the one on each card before you ship.
What a replacement has to do
- Record audio from the macOS microphone, run speech-to-text inference with a local/hosted Whisper-compatible model, display live and final transcripts in a simple UI, export transcripts (TXT/SRT) and basic timestamps.
What it still won’t have
- Polished macOS native UX and system integration (installer, auto-updates)
- Commercial support and warranty
- Any proprietary convenience features or packaged optimizations the vendor may provide
What remains hard
- Product polish and ongoing maintenance
First-year cost
No published price
MacWhisper does not publish a price we could read, so there is nothing to compare against. What building costs is below.
Money you would actually spend
Time you would spend
—
What you would spend
What we assumed
The verdict above measures whether you could build it. This one is only about money.
Runnable build prompt
Build a minimal macOS local speech-to-text app using Swift (AppKit) or Electron + a Rust/Python backend. Include: 1) a microphone capture service that captures audio frames and saves recordings; 2) a model inference adapter that can call a local Whisper-compatible binary or a hosted API (configurable); 3) a simple UI that shows streaming partial transcripts and final transcripts, allows pausing/starting recording, and exports TXT and SRT; 4) persistent settings for model choice and export path. Out of scope: code-signing, App Store packaging, commercial auto-update server. Require: error handling for microphone access and model failures, unit tests for the inference adapter, and an end-to-end smoke test that records 5s and produces a transcript.
How we checked
How the score was reached
- Build verdict base78
- An open-source build was found+5
- 3 cited sources+3
- Evidence score86
The base comes from the verdict. Everything under it is a check that either happened or did not, and each one is a fact frozen in this record rather than a judgement made at render time — so the same evidence always produces the same number.
How scoring works →Cited sources · 3
Every page the run actually retrieved.
- official productMacWhisper product page
- open sourceKoljaB/RealtimeSTT (repo description)
- open sourceufal/whisper_streaming (repo description)
Integrity checks
What held up, and what did not.
