Skip to content

Releases: lewiswigmore/macOS-dictate

Release list

v0.2.0

v0.2.0 Latest
Latest

Choose a tag to compare

View all tags
@github-actions github-actions released this 11 Jun 17:51
v0.2.0
f9b2027
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode. 
dictate v0.2.0 — distil-medium.en default + opt-in NVIDIA Parakeet ba…
Assets 7
Loading

dictate v0.1.0

dictate v0.1.0

Choose a tag to compare

View all tags
@lewiswigmore lewiswigmore released this 25 May 21:37
v0.1.0
a62671e
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

Added

  • Initial public release.
  • Hotkey-driven recording (hold / tap / double-tap state machine).
  • Streaming faster-whisper ASR with silero VAD.
  • Cleanup pipeline (off by default for privacy): opt-in Ollama local
    cleanup and opt-in OpenRouter cloud cleanup, toggled per backend in
    Settings.
  • Per-app vocab presets: code, work, personal, projects.
  • Voice commands: scratch that, new line, new paragraph, spell that ….
  • Replacement dictionary layer (config/vocab/replacements.txt).
  • Regex-driven secret redaction layer.
  • Synthetic Cmd+V insertion with always-restore clipboard.
  • Secure-input guard (refuses to paste into password fields).
  • Click-through HUD + rumps menubar with status badges.
  • First-run onboarding wizard.
  • dictate doctor diagnostic subcommand.
  • dictate --version and dictate --dry-run CLI flags.
  • dictate:// URL scheme handler.
  • AppleScript dictionary (assets/dictate.sdef).
  • Loopback-only WebUI for reviewing history (dictate-web, default port 47843).
    • Stable SHA-256 entry IDs.
    • CSP, nosniff, referrer-policy headers.
    • fcntl advisory file locking for concurrent writes.
    • Empty state, dark mode (prefers-color-scheme), mobile-responsive nav.
    • Keyboard shortcuts and a11y landmarks.
  • macOS conflict detection (built-in Dictation, Voice Control, other dictation apps, hotkey interceptors).
  • Runtime resilience: AVAudioEngine reset on sleep/wake; audio route-change rebuild; Ollama circuit breaker; disk-full callback; invalid-YAML recovery.

Documentation

  • README, CONTRIBUTING, CODE_OF_CONDUCT, SECURITY (via GitHub Security Advisories), CHANGELOG, ROADMAP, FAQ.
  • THREAT_MODEL and THIRD_PARTY_NOTICES.
  • docs/architecture.md with Mermaid diagrams.
  • docs/pre-launch-checklist.md.

Tooling

  • MIT license.
  • CI on macos-14 with Python 3.11 + 3.12 (ruff + pytest + advisory pip-audit + bandit).
  • pyproject.toml with hatchling build backend and [project.scripts] for dictate and dictate-web.
  • justfile, py.typed (PEP 561), pre-commit config, CODEOWNERS, dependabot.
Loading