Let your microphone decide. While you are quiet the video leads, from the moment you speak you lead, and the first and last seconds go to your intro. In ReactionCut that cut is already on the timeline when you open the last step, and every piece of it stays editable.
Three moments, and your microphone picks between them. The colours are the ones the timeline uses.
Before it cuts, the editor reads your whole microphone track. The quietest part is your room, the loudest part is your voice, and the line between quiet and talking goes just above the room. A fixed number would be too eager in a noisy room and deaf to someone who talks softly. If the recording has no clear quiet and loud parts, it falls back to a standard value you can move yourself.
Every stretch of talking is extended by a quarter of a second into the quiet on either side, so the camera is on you before the first word and stays for the last one. The cut never lands in the middle of a syllable.
They sit under Cut on the left of the last step, named exactly as below. Most projects never need them; they are there for the one that does.
One switch above the sliders, and the two answer different videos.
Quiet stretches are shortened to the length on Trim long silence to, and pieces shorter than Drop clips under are dropped. The finished video is shorter than your recording. It suits highlights, trailers and most reactions to long footage.
Nothing is dropped and nothing is shortened: the video is exactly as long as your recording, and only the layout changes when you talk. It suits full episode and movie watch-alongs, where viewers want to watch along with you.
A silence remover built for talking-head videos deletes every pause and closes the gap. On a reaction that is the wrong tool: the stretch where you are silent is exactly where the video plays, and deleting it deletes the thing you are reacting to.
So ReactionCut does two different things with quiet. Short pauses in your talking are ignored (Ignore pauses under). Long quiet stretches are shortened, not removed: with Trim pauses on, a two-minute stretch where you only watch becomes the last few seconds before you speak again, so the moment you react to keeps its build-up. Set Trim long silence to low for a tight cut, or high to let scenes breathe.
The cut is only as good as the voice it listens to.
Through speakers your microphone hears the video, and a loud scene looks like you talking.
Add your microphone as a separate file on the Files step, and the cut follows your voice only.
Whispered asides sit close to the room noise. If they matter, lower Silence is below a few steps.
The cuts are measured on your camera and placed on the video through the sync. A wrong sync puts right cuts on the wrong scene.
Setting up the recording itself: how to record a reaction video with OBS. Lining the two files up: how to sync a reaction video.
No. It does not transcribe anything. It measures how loud your microphone track is from moment to moment and decides where you are talking from that. That is also why it works in any language.
Yes, but less precisely: a loud scene in the video can look like you talking. Record with headphones, or add your microphone as its own file on the Files step, and the cuts follow only your voice.
Trim pauses shortens every long quiet stretch to a length you set, down to very short. It does not delete them outright on purpose: in a reaction, the stretch where you are quiet is where the video plays, which is the thing you are reacting to.
Change that clip. Pick it on the timeline and give it another look with the number keys, drag its edge, split it or delete it. Moving the sliders rebuilds the whole cut, and the editor asks first if you have already changed it by hand.
No. The credit is spent once, when a watermark-free generation finishes. Cutting, re-cutting and saving that video again are free, as is everything with a watermark.
Open the editor, add two recordings or the demo project, and the cut is waiting on the last step.
Try ReactionCut free