Presspeech for Mac
Push-to-talk dictation tool with fast on-device speech recognition and no cloud processing.
By rcourtman. Free and open source, MIT licence.
Presspeech at a glance
- Version
- 0.3.9
- Released
- 26 September 2026
- Download size
- 8.5 MB
- Works on
- Apple silicon and Intel
- Licence
- MIT
- Category
- Audio and Music
- Developer
- rcourtman
- GitHub stars
- 34
About Presspeech
Fast, private push-to-talk dictation for Apple Silicon Macs — local Parakeet TDT v3, no cloud or telemetry.
How to install Presspeech on a Mac
- Get OpenDMG. It is a free app store for open source Mac apps. Download OpenDMG.
- Search for Presspeech and press Get. OpenDMG downloads version 0.3.9 from rcourtman's own GitHub release.
- Open it. The download is checked against its SHA-256 fingerprint and its signature is checked on your Mac before it lands in Applications.
Updates to Presspeech show up in OpenDMG and install with one button. No account is needed.
Prefer to do it by hand? The developer's file is on GitHub: Presspeech.zip (8.5 MB).
What is new in Presspeech 0.3.9
# Presspeech 0.3.9 > **Before opening macOS 0.3.9:** On a clean install without a complete model > cache, Setup asks you to choose **Download Model** before the first network > transfer; **Set Up Later** defers it. Existing installations retain automatic > startup, so a missing model may still download on launch. This release > removes inherited Hugging Face account tokens from Presspeech's model > requests, but valid inherited proxy settings still route them; check the > proxy's trust before a download. Dictation audio and transcripts are not sent > with model requests. Read the > [macOS model-download privacy guide](https://rcourtman.github.io/presspeech/install.html#model-download-privacy) > before deciding to launch or clear a working cache. ## Setup accessibility - A clean install now waits for an explicit **Download Model** choice before starting the first ~500–600 MB speech-model download. Setup also shows the estimated free space needed for download and CoreML preparation before the choice. **Set Up Later** defers it; an approved download resumes on later launches. Existing installations keep their automatic startup, and a cached model loads without another prompt. - Routine model-progress and permission-label updates in Setup Checklist now update the existing accessible elements instead of replacing the full view …
More Audio and Music apps
-
MonitorControlAdjust external monitor settings
-
FluidVoiceVoice-to-text app with AI enhancement
-
eqMacSystem-wide audio equalizer and volume mixer
-
VoiceInkReal-time speech-to-text app
-
LyricsXDisplay lyrics for iTunes/Spotify
-
FreeFlowPush-to-talk speech-to-text tool that inserts polished transcripts into any app