[HN Gopher] Show HN: Privacy-First Voice-to-Text for macOS
___________________________________________________________________
Show HN: Privacy-First Voice-to-Text for macOS
Author : irqlevel
Score : 20 points
Date : 2025-09-27 13:45 UTC (9 hours ago)
(HTM) web link (github.com)
(TXT) w3m dump (github.com)
| leakycap wrote:
| This looks awesome, there are a lot of Whisper-based tools on
| macOS that have become paid for basic features
|
| I'll definitely try this out
| irqlevel wrote:
| Thank you -- any feedback is appreciated. I want to be useful,
| not rich, so I made it open and free for everyone
| Leftium wrote:
| Also check out: https://handy.computer
|
| - (Show HN discussion: https://hw.leftium.com/#/item/44302416)
|
| WhisperClip is cool because it transcribes as you are
| recording. Most other tools I've tried wait for the recording
| to finish before any transcription starts.
| seemaze wrote:
| As someone who doesn't use dictation in any form, I'm curious how
| this compares to the built-in dictation functions of macOS or
| iOS?
| youniverse wrote:
| I believe audio is normally processed on device but transcripts
| can be sent to apple.
|
| Even apps like superwhisper I'm looking to ditch because during
| an update they swapped my local models to cloud models which is
| a crazy breach of trust.
|
| Very happy to see this stuff!
| malshe wrote:
| I am definitely gonna try this out. I use dictation daily so this
| looks really useful.
| malshe wrote:
| Just now tried it. It's really good! I will keep using it more
| today to see if it performs well consistently.
|
| I had a small issue while onboarding. The app asked for
| microphone access, which I granted. It showed the task was
| completed. But the blue Next button stayed grayed out. I forced
| quit the app and then it all worked out fine from that point
| onward.
|
| I am using M4 Pro Mac Mini running MacOS 15.6.1
| irqlevel wrote:
| Thank you for the feedback. I'll work on reproducing the
| issue and will do my best to fix it.
| Leftium wrote:
| This also happened to me.
|
| And the overall onboarding/setup was clunky compared to other
| apps
|
| - SuperWhisper (probably has the best onboarding)
|
| - Handy
|
| - Whispering
|
| ---
|
| I think it could be improved by at least removing optional
| steps like downloading the LLM for transformations.
| irqlevel wrote:
| Thank you for the feedback. I will improve the onboarding
| and setup in the next app release.
| irqlevel wrote:
| This issue has been reproduced and fixed in version 1.0.44,
| now available on the website. Note that when you update the
| app, the models will be recompiled, so the app may briefly
| hang during recording. Each new app version - new hash - AI
| (MLX) code recompilation.
| Leftium wrote:
| The demo video[1] shows the transcription starts before recording
| is stopped, but I couldn't recreate this with the actual app. The
| transcription would not show after a minute of recording, but it
| was all there once I hit stop.
|
| Was this changed since the video was made? Or is there a setting?
|
| [1]: https://youtu.be/6n0jY7k-iz4
| irqlevel wrote:
| Sorry about that -- this feature is no longer available due to
| its cloud-based nature, but I plan to bring it back. The video
| was recorded before open-sourcing, so I'll remove it to avoid
| confusion.
|
| Update: Replaced the site video with a demo showing the actual
| app behavior.
___________________________________________________________________
(page generated 2025-09-27 23:01 UTC)