Skip to content

DictateAI 0.2.0

Choose a tag to compare

View all tags
@ali050786 ali050786 released this 18 Oct 08:02
· 56 commits to main since this release
v0.2.0
6dc79c1
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Nothing else to install. The speech engine (WhisperKit) and the AI cleanup engine (MLX) now ship inside the app, so you don't need Homebrew, uv or Terminal.

Install

  1. Open DictateAI-0.2.0-arm64.dmg and drag DictateAI into Applications.
  2. Open DictateAI. When macOS says it could not verify the app, click Done, then go to System Settings → Privacy & Security and click Open Anyway (the app isn't notarized yet).
  3. Allow Accessibility, then the microphone on your first dictation.

You're dictating about a minute after you open it, while it fetches its speech model once. AI cleanup is optional: turn it on in Settings, and it downloads in the background while you keep dictating.

Also in this release

  • Settings shows progress while models download, and slow connections no longer time out after 5 minutes.
  • MLX stays the cleanup engine: llama.cpp was benchmarked and was 1.5–2× slower for non-English dictation.
Download sizes
  • App: 274 MB (was 127 MB; it now includes the engines)
  • Speech model, on first launch: about 500 MB
  • AI cleanup, only if you turn it on: 1.7 GB for English, 4.3 GB for other languages

Requires a Mac with Apple Silicon (M1 or later).

Assets 3
Loading