Turn a long video into short clips people watch

Under 60 seconds is the safe answer - it fits every platform and sits inside every sweet spot. Reels and TikTok often do best at 15–45 seconds; YouTube Shorts tolerates a little more since viewers arrive expecting substance. Length should follow the moment, not the other way round.

No. Clippingnet cuts on the time axis only - it extracts your segment at the original aspect ratio. To reframe for vertical, use CapCut, your phone's built-in editor, or the crop step most platform uploaders offer. It takes about a minute per clip.

No, and you generally don't need it to. YouTube, Instagram, and TikTok all auto-generate captions at upload now. Turn them on and check the names and jargon, which is where auto-captioning consistently gets things wrong.

No - there's no highlight detection here. You choose the moments. AI clipping tools exist and are useful at volume, but they upload your video and often pick the wrong forty seconds. This tool assumes you know your content better than a model does.

As many as there are good moments - usually three or four per hour of well-made long-form. The source stays loaded between exports, so cutting several in one session is quick. There's no cap on exports.

You can mark in/out points on any YouTube link and get a timestamped share link plus an auto-stopping embed. You can't download a file - YouTube doesn't permit browser-side downloads. For a file, use your own upload or a local copy.

Yes. Local files are read from disk rather than uploaded, so length and size barely matter - a three-hour, 8 GB stream VOD opens in about a second. The only limit is what your browser can decode.

The browser records your selection while it plays, which is how in-browser capture works without a server. A 45-second short takes about 45 seconds. Preview first so you don't spend that time on a cut you'll redo.

WebM with full audio at 8 Mbps. Every major platform accepts WebM uploads. If you need MP4 - for instance because your editing app prefers it - convert in one step with ffmpeg -i short.webm -c:a aac short.mp4.

Later than feels natural. People preamble before making a point - "so I think that, basically…" - and that wind-up is where short-form viewers leave. Start on the first word of the actual claim, which is often ten seconds after where you'd instinctively cut.

Usually just the answer, unless the question is short and the answer is meaningless without it. A clip should stand alone for someone with no context - if you need three seconds of setup, keep it; if you need thirty, pick a different moment.

You can, but it usually performs worse than either alone. Cut them separately and post on different days - two chances at attention beats one diluted attempt, and each clip stays tight.

1080×1920 is the standard for vertical. The export preserves your source resolution, so a 1080p horizontal source gives you plenty to crop from. Anything below 720p source will look soft once cropped to vertical.

Only with permission. Many creators actively want clips and grant permission through official clipping programs - that's the clean path. Reposting without any permission, especially monetised, is how channels get claims and strikes.

The best hook is usually the moment itself, starting at its most interesting sentence. Bolted-on "wait for it" intros mostly cost you the first two seconds, which is exactly where you can least afford to lose attention.

Slightly, since the segment is re-encoded during capture. At 8 Mbps it's hard to see, and social platforms re-encode everything on upload anyway - their compression will do far more to the image than this step does.

Yes - the timeline is touch-friendly and works in mobile browsers. Since reframing and captioning usually happen on your phone anyway, cutting there too can keep the whole workflow on one device.

Nothing is uploaded - your file is read from disk and cut locally. If you're pulling teasers from an episode that hasn't published, it never reaches a server you don't control. Cloud repurposing tools can't offer that.

Audio and video are captured together from the same playing stream, so sync is preserved automatically. If your source already had drift, the clip inherits it - fix that in the source before cutting.

If the long video is cleared, the clip generally is too - but platform content-ID systems check clips independently and can flag music that passed elsewhere. If a clip gets claimed, the usual fix is choosing a moment without the music bed.

Consistency matters more than volume. A reliable two or three a week beats eight in one day and nothing for a month. Cutting four clips from a single episode gives you a fortnight of scheduled posts.

Yes, in the caption or pinned comment - that's the whole point of repurposing. On YouTube you can link Shorts directly to the source video, which makes the funnel explicit rather than hoping people search.

Then you're cropping, and something leaves the frame. A centre crop works when the subject is centred; a blurred-background fill keeps the whole frame at the cost of screen space. Both are one step in any editor - there's a ready-made FFmpeg command for each above.

Once the stream has ended and the VOD exists, yes - load the file or paste the VOD link. While a stream is still live, use the platform's own clip feature; there's no fixed timeline to mark points on yet.

None. There's no credit system, no monthly cap, and no watermark on any of them. Since your own browser does the work rather than a server, there's nothing to meter.