logoalt Hacker News

Show HN: OpenWhisper – free, local, and private voice-to-text macOS app

28 pointsby rwu1997yesterday at 6:51 PM8 commentsview on HN

I wanted a voice-to-text app but didn't trust any of the proprietary ones with my privacy.

So I decided to see if I could vibe code it with 0 macOS app & Swift experience.

It uses a local binary of whisper.cpp (a fast implementation of OpenAI's Whisper voice-to-text model in C++).

Github: https://github.com/richardwu/openwhisper

I also decided to take this as an opportunity to compare 3 agentic coding harnesses:

Cursor w/ Opus 4.6: - Best one-shot UI by far - Didn't get permissioning correct - Had issues making the "Cancel recording" hotkey being turned on all the time

Claude Code w/ Opus 4.6: - Fewest turns to get main functionality right (recording, hotkeys, permissions) - Was able to get a decent UI with a few more turns

Codex App w/ Codex 5.3 Extra-High: - Worst one-shot UI - None of the functionality worked without multiple subsequent prompts


Comments

PhilippGilletoday at 7:36 AM

That's a very similar name as https://github.com/OpenWhispr/openwhispr

Which is a voice to text app as well.

stavrostoday at 2:32 AM

I really like https://handy.computer for this.

show 1 reply
kristianptoday at 6:29 AM

MacWhisper has been around for a while and it says no data leaves your machine.

https://goodsnooze.gumroad.com/l/macwhisper

zenapollotoday at 1:37 AM

I thought i had tried this but no it’s a massively crowded app name. I’ll check this out soon, but honestly this one has been great.

https://github.com/Starmel/OpenSuperWhisper

Though it seems like the dev stopped work on it

dcsantoday at 12:48 AM

how does it compare to wispr flow?

which beats the pants off android or iphone transcription...

eductiontoday at 5:31 AM

Requiring microphone access is a dealbreaker. I transcribe audio files I’ve recorded on other equipment. You don’t need my mic for that.

I clicked hoping it would leverage whisper’s translation capabilities as well, since MacWhisper does not do that (it leverages Apple’s subpar built in translation). It doesn’t do that either.

(Maybe you’re using this for dictation? That’s a very specific subset of transcription. I’d suggest using a name that leans into that.)

umtksayesterday at 9:55 PM

the video on github is not working