Jump-Cut Remover
Cuts silences, "ums" and pauses out of video recordings automatically. Also includes a waveform editor for manual fine-tuning.
What it does
Use this to turn raw vlogs, training videos, podcast video recordings or YouTube footage into polished content. Typically trims 20-50% of the raw runtime and creates a dynamic pace.
How to use
- Drag video files into the list.
- Pick a Quick Preset (Podcast, Tutorial, Vlog, Aggressive, Conservative).
- Click Analyse First File.
- Review the detected silence regions on the waveform.
- Toggle each region to CUT or KEEP as needed.
- Click Run.
You get one cleaned-up copy per video.
Quick presets
| Preset | What for |
|---|---|
| Podcast | Speech-focused, preserves natural pauses. |
| Tutorial | Balanced for explanatory content. |
| Vlog | Fast pace, even short pauses go. |
| Aggressive | Maximum tightening, only speech remains. |
| Conservative | Light cleanup, preserves natural flow. |
Fine controls
- Silence Threshold (dB): What level counts as silence. -40 dB is standard.
- Min Silence Duration (ms): Targets silences longer than this. 400 ms is standard.
- Safe Padding (ms): Leaves a bit of audio at the edge of each silence so cuts are not jarring. 100 ms is standard.
Waveform editor
After analysing the first file, the waveform appears with detected silences colour-coded. Click each region to toggle CUT or KEEP. The preview player lets you listen before deciding.
- Zoom in/out: Ctrl + Scroll or the zoom buttons.
- Mark all regions: Bulk CUT or KEEP.
Standard vs AI mode
- Standard mode: Fast, detects silence by audio level.
- AI mode: Uses Whisper to also catch filler words like "umm" and "uh". CUDA GPU recommended.
Examples
Tighten a vlog: Add the video, Vlog preset, analyse, review, run. Cuts the runtime roughly in half.
Clean up a podcast video: Add the video, Podcast preset, AI mode (if available), run. Fillers go.
Shorten long pauses in a training video: Add the video, Tutorial preset, threshold -42 dB, min 800 ms, run.
Quick draft: Add the video, Aggressive preset, run without reviewing. Fast result, no listening required.
Watch out
- AI mode needs Whisper installed and a CUDA GPU is recommended. On CPU it is 10-30x slower.
- Very aggressive settings break the natural rhythm of speech.
- Manual edits on the first file only apply to that file. Other files in batch use the automatic detection.
- Do not use on videos with music.
- The change is permanent, keep the original.
License
Standard mode is available in the Free tier with a monthly cap. AI mode is Ultimate only.