github jatinkrmalik/vocalinux v0.2.0-alpha

7 months ago

🎉 Vocalinux v0.2.0-alpha // First Public Alpha Release

The seamless voice dictation system for Linux is here!

Vocalinux brings hands-free voice typing to your Linux desktop with powerful offline speech recognition. Whether you're writing documents, coding, or just prefer speaking over typing, Vocalinux makes it effortless. Anywhere in your OS.


✨ Highlights

  • 🎤 Whisper AI as Default Engine - High-accuracy speech recognition powered by OpenAI's Whisper, running 100% offline
  • ⚡ One-Command Installation - Simply curl | bash and you're ready to go
  • 🔒 Privacy First - All processing happens locally, no data leaves your machine
  • ♿ Accessibility Focused - Built for users who need voice input for daily computing

📦 Installation

curl -fsSL https://raw.githubusercontent.com/jatinkrmalik/vocalinux/main/install.sh | bash

That's it! The installer handles everything including downloading the Whisper tiny model (~75MB).


🚀 What's New in v0.2.0-alpha

Speech Recognition

  • Whisper AI Integration - Full support for OpenAI's Whisper models (tiny, base, small, medium, large)
  • Whisper tiny as default - Optimized out-of-box experience with the tiny model pre-downloaded
  • VOSK fallback - Lightweight offline recognition as backup engine
  • Voice Activity Detection - Smart silence detection to auto-stop recognition
  • Model size selection - Choose accuracy vs speed tradeoff in settings

User Experience

  • Double-tap Ctrl to toggle - Natural keyboard shortcut to start/stop dictation
  • System tray indicator - Visual feedback with microphone status icons
  • Audio feedback - Sound cues for start/stop/error states
  • Settings dialog - GTK-based configuration UI for all options
  • Wayland support - Works on both X11 and Wayland (with xdotool/ydotool)

Installation & Setup

  • One-command curl install - curl -fsSL ... | bash works seamlessly
  • Non-interactive mode - CI/automation-friendly with -y flag
  • Pre-downloaded models - Both Whisper tiny and VOSK models installed during setup
  • Desktop integration - App icon, .desktop file, and system tray support
  • Clean uninstall - uninstall.sh removes everything including curl-installed files

Developer Experience

  • Comprehensive test suite - 80+ tests with pytest
  • CI/CD pipeline - GitHub Actions for linting, testing, and releases
  • Pre-commit hooks - Black, isort, flake8 for code quality
  • Debug logging - Built-in logging UI for troubleshooting

📋 Full Changelog

Features

  • feat: Make Whisper tiny the default engine and download during install (#45)
  • feat: Alpha Launch v0.2.0 - Complete UX Overhaul (#39)
  • feat: Enhance Whisper Support & Improve Settings UI (#38)
  • feat: Package VOSK model with installation for faster first-run (#32)
  • feat: Add comprehensive logging system with UI (#31)
  • feat: Comprehensive bug fixes and feature improvements (#30)
  • feat: Custom Icons Implementation and Package Restructuring (#21)
  • feat: Simplify Keyboard Shortcuts to Double-tap Ctrl (#9)
  • feat: Add audio feedback system with sound notifications
  • feat: Wayland compatibility with XWayland fallback

Fixes

  • fix: Install/uninstall scripts for curl | bash usage (#41, #40)
  • fix: Read version from file instead of importing (#39)
  • fix: CI Test Failures and Speech Recognition Configuration (#24)

Documentation

  • docs: Update all markdown files for new install flow
  • docs: Add Codecov integration for coverage reports (#20)
  • docs: Standardize on WAV audio format and improve attribution (#19)

Infrastructure

  • ci: Unify CI workflows into single pipeline with path-based filtering (#28)
  • ci: Add pre-commit hooks to match GitHub CI pipeline (#7)
  • ci: Test structure reorganization and improvements (#5, #6)

⚠️ Known Limitations (Alpha)

  • X11/Wayland text injection - Some applications may not receive text correctly on Wayland
  • First-run Whisper load - Initial model load takes a few seconds
  • CPU-only inference - GPU acceleration not yet implemented
  • English only - Multi-language support planned for future releases #13

🙏 Contributing

We welcome contributions! See CONTRIBUTING.md for guidelines.


📄 License

GPL-3.0 - Free and open source forever.


Thank you for trying Vocalinux! 🐧🎤

Don't miss a new vocalinux release

NewReleases is sending notifications on new releases.