Why Valorant is hard to crop and easy to read
Valorant spreads its information across the whole frame. The minimap top left, the round timer and score top centre, your abilities and ammo along the bottom, your teammates down the side. A clipper that crops to the centre of the frame keeps your crosshair and loses everything that tells a viewer what round it is, who is alive and why the clutch was a clutch.
At the same time, Valorant is one of the easiest games to clip from the voice track, because the game is played out loud. You call positions, you count the enemies left, you say the thing you are about to try. That running commentary is a transcript of the round, and it is what we read to choose the moments.
What the stacked layout does with a round
The gameplay takes the lower band of the vertical frame at full width and its own shape, scaled down rather than trimmed. Nothing is cropped away, so the minimap, the timer, the economy and the killfeed are all still on screen, smaller but present. Your facecam is found wherever your overlay puts it, cut out and enlarged to fill everything above the game.
The captions sit on the lower part of your facecam, over your shoulders, well clear of the round. The hook line sits at the very top. A viewer sees your face, your words and the round in one frame, which is the whole reason the moment was worth posting.
Clutches start before the last kill
A one versus four is not interesting at the fourth kill. It is interesting from the moment your last teammate drops and you say, out loud, that you are going for it. Clippers that key on the shouting find the fourth kill and open there. Because we read the transcript from start to finish and choose stretches with a beginning, a clip from a Valorant session here tends to open on the call, with the kills as the payoff.
Any agent, any map, any rank
Nothing in the pipeline is Valorant specific. It reads your voice and finds your facecam, so a new agent or a map it has never seen behaves exactly like one it has. The one recording it cannot use is a silent one: a spectator recording with no voice track has nothing to read.
Two hours in, a few Shorts out, posted from here
A pasted YouTube link can be up to four hours, and on the free tier each of your first two videos can be up to two hours, which is a full ranked session. When the clips are ready we email you. Give one a title and it uploads to your own YouTube channel as a Short, private, unlisted or public, only when you press the button.
The clutch you called is in your last VOD
Point us at the recording and the rounds you talked through are cut, captioned and waiting with your facecam still in frame. The first two videos are free.
Frequently asked questions
›Does it detect aces and clutches from the game?
No. It reads what you said. If you called the clutch, reacted to the ace or talked through the round, that is found. A round you played in silence is invisible to it, which we would rather tell you than imply otherwise.
›Will the minimap and timer be readable on a phone?
They are kept, at a reduced size, because the whole frame is scaled into the lower band rather than cropped. They are small but present, which is better than the usual outcome of a centre crop, where they are simply gone.
›What if I record without a facecam?
Then there is nothing to stack and the clip is reframed to 9:16 the ordinary way, keeping the action centred. The stacked layout only comes into play when there is a facecam to keep.
›My VOD is on Twitch, not YouTube. Can I use it?
We fetch from YouTube, not Twitch. If you export your VODs to YouTube, sign in or paste the link. If not, download the VOD from your Twitch dashboard and upload the file here.
›Can other people's voices in team chat be captioned?
Yes, if they are in the recording. The transcript covers the whole voice track, and the captions follow whoever was speaking.