Skip to content

Video to Shorts converter

Turn a long video into vertical 9:16 clips for Shorts, Reels and TikTok, with auto captions, a crop that follows the speaker and no watermark. It all runs on your device.

I want Make the shorts

Settings appear once you add a video.

Fit it into 9:16
Captions
Caption look
Advanced 3
Fill the screen from the
Show the headline

Done on your device No upload, no queue, no sign-up.

Bytes sent anywhere: 0 B

Works offline after you’ve used it once.

How to use it

Your long video goes into this browser tab, is cut and reframed into vertical 9:16 clips on your device, and comes back as several shorts. 0 bytes uploaded

  1. Add your file

    Drop a long video. Its liveliest moments are suggested as clips: preview them, move them, remove them or add your own.

  2. Check the settings

    Choose how the picture fills the 9:16 frame — the whole video, or a crop that follows the speaker — and a look for the automatic captions.

  3. Save the result

    Press Make the shorts. Each clip is made at 1080×1920 on your device and saved as an MP4, with no watermark and no limit.

Why it doesn’t upload

Clip services want your whole video on their servers first: an hour of podcast, a private stream, a client’s webinar. Here the captions are written, the faces found and the clips rendered in this tab, so the video never leaves your device — and there are no minutes to buy.

How to check for yourself

Worth knowing

  • Captions and face tracking run on your device. Captions come from Whisper, a speech-recognition model, and face tracking from MediaPipe’s face detector. The speech model downloads once (about 77 MB, from Hugging Face, where it’s published) and the browser keeps it; your video is never sent anywhere.
  • Writing captions takes roughly as long as the clips play on a laptop, longer on a phone. Check them before you post: names and unusual words are where speech recognition slips.
  • Follow the speaker follows the main face in each shot. It can’t tell who is talking, so when two people sit far apart it stays with one of them; choose Whole video to keep both in the picture.
  • Clips are suggested by how loud and lively the sound is, not by what is said, so check them with the preview buttons.
  • It works on files, not links: it can’t fetch a video from YouTube. For your own videos, download the file from YouTube Studio first. MP4, MOV, MKV and WebM are read directly; convert other formats first.

Whole video, or following the speaker

A 16:9 picture shrunk to fit a 9:16 frame fills barely a third of it. Whole video keeps the entire picture and fills the rest with a soft, darkened copy of the same video, so the clip still feels full-screen — the right choice for two-person podcasts and wide shots, with room above and below for a headline and captions.

Follow the speaker fills the whole frame instead, cropping in on the main face. The crop is planned with the whole clip in view, so it stays perfectly still while someone talks, pans smoothly when they move, and jumps with the picture at every camera cut — no wobble, no sliding across the room.

Captions that keep people watching

Most people scroll short videos with the sound off, so the captions do the talking. Highlight shows three words at a time and lights up each one in yellow as it’s said; Box puts the spoken word on a colored box; Pop shows one or two big words at a time that pop onto the screen; Simple shows whole sentences, for interviews and explainers. All four use the same heavy, outlined type that stays readable over any picture.

Got a YouTube link?

Links can’t be used: fetching a YouTube video needs a server in between, and YouTube’s terms only allow downloads through YouTube’s own features. Your own uploads are easy to get back, though: in YouTube Studio, open Content, hover over the video, click the three dots and choose Download. For every video on a channel at once, Google Takeout exports them all.

Lengths and sizes that work everywhere

Every clip is 1080 × 1920 pixels, H.264 video with AAC sound in an MP4 — the format YouTube, Instagram and TikTok all recommend for vertical video. Clips of 15 to 60 seconds are the safe range for all three; YouTube Shorts take up to three minutes. See the YouTube Shorts, Instagram and TikTok size guides for the rest.

Questions

How do I turn a long video into Shorts?

Drop the video here: clips are suggested from its liveliest moments. Preview them, move or remove any, add your own with Add a clip from here, choose the framing and captions, and press Make the shorts. You get one vertical 1080×1920 MP4 per clip, ready for YouTube Shorts, Instagram Reels or TikTok.

How are the captions written?

By Whisper, a speech-recognition model that runs in your browser. It listens to each clip and times every word, so the captions can light up each word as it’s said. Pick a look — Highlight, Box, Pop or Simple — and they’re burned into the video. Nothing is sent to a server; only the model is downloaded, once. You can also use your own subtitle file.

Can the crop follow the person speaking?

Yes: choose Follow the speaker. Each clip is watched first to find the faces and the camera cuts, then the crop is planned like an editor would: it holds still while the person stays in place, pans smoothly when they move, and cuts with the picture when the video cuts. With nobody on screen, it keeps the middle.

Can I paste a YouTube link?

No. A web page can’t fetch a YouTube video without going through a server, and YouTube’s terms don’t allow downloading videos outside YouTube’s own features. For your own videos, open YouTube Studio → Content, click the three dots next to a video and choose Download, then drop the file here.

How do I make a horizontal video vertical without cropping?

Choose Whole video: the full landscape frame sits in the middle of the 9:16 clip, over a blurred, darkened copy of itself, so nothing is cut off. The bands above and below are where the headline and captions go.

Is this an alternative to OpusClip?

For making the clips, yes, and without the limits: AI clip services upload your video and give free plans a monthly allowance and a watermark. Here captions and face tracking run on your device, there’s no watermark and no limit on how many clips you make. What it doesn’t do is judge which moments are most interesting by what is said — you choose, with suggestions from the sound.

Read more about this

Next steps

Often used before or after Video to Shorts. These run on your device too.