No one is talking.It still finds the moment.
A clipper that only reads the transcript has nothing to work with the second the words stop. Cutlist scores the sampled frames and the audio peaks directly, so a play, a reaction, a visual change registers as the moment it is - no speech to transcribe, no transcript required. This is the footage it was built for.
It measures what it hearsand sees.
The two peaks below are the moments. Motion spikes as the thing happens and the crowd arrives a beat later - a shape a transcript records as nothing at all.
- motion
- audio
- scene cut
- nominated
The two peaks are the plays. Motion spikes as it happens and the crowd arrives a beat later, which is what separates a score from a camera pan — and the commentator is often still mid-sentence about the last drive, so a transcript alone puts the clip in the wrong place.
Illustrative shape · your own curves are measured, not drawn
The audio peaks
Energy and spectral texture across the whole file. A crowd, a whistle, a drop, a reaction on the mic - each is a peak the engine measures directly, with no words involved.
The motion and the cuts
Where the frame moved and where the shot changed, sampled across the span. This is what a vertical crop follows, and what tells a real moment from a slow pan.
The same rubric
A moment still has to open well, stand alone and land somewhere. That judgement is made off the frames and the sound instead of the transcript - the score is a read of the footage, not the words.
Footage the wordswere never going to cover.
If the interesting second is something you watch or hear rather than something someone says, this is the case the engine was designed around.
- Gameplay
- A long session where the moment is a play, a clutch, a reaction on the mic - not a sentence.
- Sport
- A wide broadcast with crowd and whistle and maybe no commentary at all. The loudest signal is the stadium.
- B-roll and montage
- Establishing shots, drone passes, cutaways. Scored on the picture, because there is nothing else.
- Timelapse and process
- A build, a cook, a paint. The interesting second is a visual change, and the frames are where it lives.
- Performance
- Music, dance, a set. Audio energy and the cuts carry it where a lyric transcript would only mislead.
- Ambient and nature
- Wildlife, weather, a room. Long stretches of nothing, then the thing you were waiting for.
What silencecosts you.
One thing genuinely changes when nobody speaks, and it is better said here than discovered in the editor.
Captions are transcribed speech. Footage with no speech ships without them, because there is nothing to write - the clip is the picture and the sound, cut and reframed, and any words you want on it you add yourself. Where a clip has some speech, that part is captioned in the language it was spoken.
Everything else works exactly as it does on footage full of talk: the ranking, the vertical reframe that follows the motion, the editor, the zip, and no watermark on any plan.