About Spurt

Spurt is a cross-platform push-to-talk dictation tool powered by whisper.cpp. It allows users to press and hold a hotkey, speak into their microphone, and then release the key to have their words automatically typed into the currently active window. This tool streamlines the process of converting speech to text in real-time, making it ideal for various productivity workflows.

Key Features:

• Cross-Platform Compatibility: Works on Windows, Linux, and macOS.
• AI-Powered Transcription: Utilizes the whisper.cpp engine for accurate speech-to-text conversion.
• Push-to-Talk Functionality: Dictate only when the designated hotkey is pressed.
• Automatic Typing: Transcribed text is directly inserted into the active application.
• Customizable Settings: Users can configure the whisper model, trigger key, key mode (hold or toggle), and maximum recording time.
• Interactive Terminal UI: Provides a dashboard for live status, pausing, and resuming dictation, as well as a configuration screen.
• Standalone Binary Option: Can be built into a single executable for easy distribution without requiring Python installation.
• Comprehensive Testing: Includes a robust test suite that mocks external dependencies for reliable testing.

Technical Details:

• Built with Python.
• Leverages whisper.cpp for speech recognition.
• Uses sounddevice for microphone capture.
• Employs pynput for global hotkey detection and keyboard simulation.
• Features a textual-based Terminal UI.
• Packaging is handled by PyInstaller for standalone executables.
• Testing is managed with pytest.