Automatic Transcription

Leverage state-of-the-art Whisper models to convert speech to text automatically and locally.

• System Checks: Harvey automatically verifies your Python environment and hardware acceleration (like Core ML) before starting to ensure peak performance.
• Model Selection: Choose a model size that matches your accuracy needs. Larger models (e.g., Medium, Large) provide better results for complex audio but require more RAM.
• Language Detection: Set a specific language for best results, or use “Auto Detect” for multilingual files.