JumpCut is a Zig-powered tool that analyzes video or audio files, detects silence, and automatically generates a seamless timeline (EDL) by keeping only the spoken sections. This helps vloggers and content creators streamline their editing process by removing silent gaps, optimizing video flow.
It uses ffmpeg libraries statically linked, so no external dependencies are needed.
This project uses a fork of Zig with bindings to FFmpeg.
FFmpeg is a project licensed under LGPL/GPL, depending on its configuration. Users should verify the compatibility of their usage with the corresponding license terms.
✅ No external dependencies needed!!.
✅ Automatic silence detection in video and audio files.
✅ Exports to editing formats like EDL (DaVinci Resolve, Premiere Pro).
✅ Efficient processing powered by Zig’s speed and safety.
✅ Supports multiple input formats (Currently MP4, with PCM 16-bit 48kHz audio).
✅ Enhances the narrative flow of vlogs and spoken content.
- Vloggers and content creators looking to speed up video editing.
- Podcasters who want to remove pauses without manual editing.
- Interview and conference editing, keeping only relevant speech.
- Clone the repository:
git clone https://github.com/DanielTowerz/jumpcut.git- Build the project:
zig build --release=fast, safe or smalljumpcut -i [inputfile] -o [outputfile] -d [decibel] -s [silence duration] -a [adjustment]
-i: Input file path. [REQUIRED]
-o: Output file path. If not provided, the output will be saved in the same directory as the input file as `[inputfile name].edl`.
-d: A negative number as decibel threshold for silence detection. Default is '-25'.
-s: A positive float number as the minimum duration of silence to be detected. Default is '1.0' seconds.
-a: A positive float number as adjustment for the start and end of the detected silence. Default is '0.3' seconds.
- Export to FCPXML.
- CLI interface with adjustable sensitivity settings.
- Support for other video and audio formats.