How do you turn a long YouTube video or stream recording into short vertical clips?

In short: To turn a long video into short clips, you find moments that make sense on their own, cut them out, reframe the horizontal picture to 9:16 vertical and add captions. You can do all of it by hand in a video editor; on a one-hour recording, finding the moments takes most of the time. Wiozen Forge takes a link or a file, transcribes the speech, scores each moment from 0 to 100 and renders the first 10 clips at 1080×1920 with captions.

Which moments make good clips?

A good clip carries one idea that someone who never saw the rest of the video can follow: a question and its answer, the joke itself, an unexpected moment in a game, the clearest line of an argument. Pieces that open with “like I said earlier” or trail off into the next topic fall flat in a vertical feed, because the viewer doesn't know what came before.

If you're doing it by hand, watch the recording at 1.5× or 2× and note the timestamps where the volume jumps, people laugh or react. Then go back to each note and set the start at the beginning of a sentence and the end at the end of one. A clip that starts mid-sentence feels cut off.

Forge does this pass with AI. It transcribes the speech first, then looks at reactions in the audio (and, on a stream from your connected channel, at how busy the chat gets) and gives every moment a score from 0 to 100. A clip runs between 5 seconds and 3 minutes, and its start and end snap to sentence boundaries in the transcript.

Turning a horizontal picture vertical

YouTube videos and stream recordings are mostly 16:9 landscape. Reels, TikTok and Shorts play full screen at 9:16 for someone holding the phone upright. There are three common ways to turn a wide frame into a tall one, and the right one depends on what's on screen.

If one person is talking, you crop the frame around their face. Every time they move or the shot changes, the crop has to follow; done by hand, this is the most tedious part. If the whole picture matters (a slide, a map, a product on screen), you shrink the frame, center it and fill the space above and below with a blurred copy of the same image. Gaming streams with a webcam usually get a split layout with the camera on top and the game below.

In Forge these three layouts are called Single face, Fit and Streamer. Single face keeps the speaker in frame and re-crops when the scene changes; if two people are talking, it splits the screen and tracks each face separately. Scenes where no face is found are centered. The Streamer layout is covered in making Twitch or Kick clips with the facecam on top.

Captions and the first three seconds

In a vertical feed, people decide within the first few seconds whether to stay, and some of them watch with the sound off. Two things help: captions that move with the speech, and a short title at the very start that says what the clip is about.

Forge shows captions word by word, in sync with the speech. There are three styles: Highlight, which marks the word being spoken, plain Classic, and Boxed on a dark background. With the hook title on, a short headline appears at the top for the first 3 seconds of the clip, and you can change its text in the editor. Transcripts and titles can be written in Turkish, English, German, Arabic, Spanish, French or Russian.

Step by step with Wiozen Forge

1. In the panel, open Clips and choose “Create clip”. Paste a YouTube, Twitch or Kick link, or upload an MP4, MOV, MKV or WEBM file from your computer. Files can be up to 2 GB, and if your connection drops the upload picks up where it stopped.
2. Pick the range to process. The default is the whole video, up to 6 hours. If you only select the part you care about, less clip time is used.
3. Choose the layout (Single face, Fit or Streamer), the caption style and the language.
4. Choose how clips are shared: download only, have the best 3 clips scheduled automatically on your connected account, or keep them waiting for your approval.
5. The job goes through audio download, transcription, moment selection and rendering. You get a notification when the clips are ready.

You can move the start and end of any finished clip sentence by sentence in the transcript. Moments that didn't make the first 10 stay in the “Other suggestions” list, and you can turn any of them into a clip with “Make clip”. Neither these extra clips nor re-rendering every clip with a new look uses clip time.

What does it cost?

In Forge, clips use the clip hours that come with your plan. The time charged is the length of the range you processed, counted when the job finishes: anything under 1 minute counts as 1 minute, and partial minutes above that are rounded down. Failed or cancelled jobs aren't charged. Downloads are unlimited.

The smallest clip plan is Streamer: $6.99 a month for 3 hours of clips. If you run out and approve it when starting a job, the extra time is charged at $0.50 per hour from your Wiozen balance when the job finishes. Up to 2 clip jobs can be open at once.

Which videos can you use?

Only process your own content or videos you have the right to use. Clipping and posting someone else's video leads to copyright trouble, and platforms can take those posts down. Forge says the same under the link field; the details are in the Content and Copyright Policy.

Frequently asked questions

How many clips do you get from a one-hour video?

Forge renders the 10 highest-scoring clips on its own. The remaining suggestions are listed on the job page, and you can make any of them without using clip time.

How long can a clip be?

A clip is between 5 and 180 seconds (3 minutes) long. You can change its start and end sentence by sentence in the editor.

Which languages do captions support?

Turkish, English, German, Arabic, Spanish, French and Russian. You choose the language when starting the job, and the transcript and clip titles are written in it.

How long are clips kept?

Clips and their transcripts are deleted within 30 days, and an uploaded source file within 3 days. Download the clips you want to keep before then.

Which plan do you need to post clips?

Downloading clips is unlimited on every plan. Automatic posting to Instagram, TikTok and YouTube Shorts starts with the Starter plan.

Sign in with your Wiozen account and start your first job from the Clips page.

Start with Wiozen