Skip to content

OpenWhisper 0.5.5

Latest
Latest

Choose a tag to compare

View all tags
@github-actions github-actions released this 14 Oct 23:23
· 1 commit to main since this release
v0.5.5
edb0875

What's new in OpenWhisper 0.5.5

  • Warm up enabled local Ollama and LM Studio models during dictation to reduce the wait after recording.
  • Keep a timestamped, copyable history for new recordings, and opt in to preview updates.
  • Retry failed AI text revisions without recording again.
  • Fall back to CPU when the optional GPU check fails.
  • Speed up vocabulary matching on long transcripts without changing results.

Neu in OpenWhisper 0.5.5

  • Wärme aktivierte lokale Ollama- und LM-Studio-Modelle bereits während des Diktats vor, um die Wartezeit nach der Aufnahme zu verkürzen.
  • Führe neue Aufnahmen mit Zeitangabe und Kopierfunktion im Verlauf; Vorschau-Updates lassen sich optional aktivieren.
  • Starte fehlgeschlagene KI-Textüberarbeitungen erneut, ohne noch einmal aufzunehmen.
  • Wechsle zur CPU, wenn die optionale GPU-Prüfung fehlschlägt.
  • Beschleunige den Wortschatzabgleich bei langen Transkripten, ohne die Ergebnisse zu verändern.

Recognition runs locally. Exact package, desktop and test scope is recorded in
Electron status and
Linux evidence.

Updating from 0.2.5

For the first transition from the native 0.2.5 application to Electron, install 0.4.1 manually
from the packages below. The original application's GUI updater has not been verified for this
host transition. The earlier 0.2.4-to-0.2.5 transition also required manual installation;
the 0.2.4 updater cannot apply 0.2.5. Older versions may need to be reopened from the application launcher
after an update. See the immutable 0.2.5 release notes
for the previous update guidance.

macOS

Download OpenWhisper-macOS.dmg, drag OpenWhisper.app to Applications, eject the image,
and open the app from Applications. A ZIP is also available. Both packages include Apple Silicon
and Intel binaries and require macOS 14 or later. Download a model before dictation.

The app is signed with the project's persistent self-signed certificate and is not notarized by
Apple. If macOS blocks the first launch, review the app's origin and use System Settings →
Privacy & Security → Open Anyway if you choose to allow it.
See release signing and
Apple's first-open instructions.

Verify the DMG with SHA256SUMS from the same folder:

grep '  OpenWhisper-macOS[.]dmg$' SHA256SUMS | shasum -a 256 -c -

Linux x86_64

Download and run the AppImage:

chmod +x OpenWhisper-Linux-x86_64.AppImage
./OpenWhisper-Linux-x86_64.AppImage

If FUSE is unavailable, run it with APPIMAGE_EXTRACT_AND_RUN=1. On Debian or Ubuntu, install
the package with:

sudo apt install ./OpenWhisper-Linux-amd64.deb

Verify downloads against SHA256SUMS. See Electron status
for the tested package scope and remaining platform coverage.

The interface is available in English and German. Dictation supports multiple spoken languages.

Assets 10
Loading