01Local video analysis: Performs scene detection, transcription, and keyframe extraction entirely on-device.
02Advanced scene detection: Identifies visual transitions rather than fixed time intervals for frame capture.
03Multi-backend transcription: Supports various Whisper implementations (mlx-whisper, faster-whisper, whisper-cpp, openai-whisper) optimized for different hardware.
04Smart presets for video types: Configures analysis parameters for screencasts, presentations, movies, interviews, and more.
05Privacy-first design: Ensures no video or audio data ever leaves the local machine.
061 GitHub stars