This tool allows you to record audio and automatically transcribe it to text using OpenAI's Whisper model.

1. Make sure you have all required dependencies installed:

   uv sync

1. Run the script:

   uv run transcription_tool.py [--model MODEL] [--list-models]

2. The tool will start and wait for your commands. You have two main hotkey combinations:

   • Ctrl+Space (All platforms)

     • Press to start recording
     • While recording, press 'Esc' to stop recording
     • The transcription will automatically begin after stopping
     • Once complete, the text will be copied to your clipboard

   • Cmd+Shift+Q (Mac) or Ctrl+Shift+Q (Windows/Linux)

     • Press to quit the application

• --model MODEL: Choose the Whisper model to use (default: "tiny")

  • Available options: tiny, base, small, medium, large
  • Larger models are more accurate but require more computational resources

• --list-models: Display all available Whisper models and exit

# Run with default (tiny) model
uv run transcription_tool.py
# Use the base model for better accuracy
uv run transcription_tool.py --model base
# List all available models
uv run transcription_tool.py --list-models

• Automatic language detection
• Direct clipboard integration
• Real-time audio recording
• Uses OpenAI's Whisper model for accurate transcription

• The first time you run the tool, it will download the Whisper "base" model
• The transcription quality depends on your audio input quality
• Make sure your microphone is properly configured before starting