AI Clip Maker｜Long-to-Shorts
Upload one long video — AI finds the best moments and cuts captioned vertical clips ready to post.
AI clip maker for long videos: upload a podcast, stream or interview and get captioned vertical clips for TikTok, Reels and Shorts. AI watches. Try it free.
Features
- Auto highlight picking——AI watches the whole video and shortlists candidates by contrarian takes, quotable lines and emotional peaks — ranked strongest first. You get the list with timestamps before anything renders.
- 9:16 reframing——Landscape footage is intelligently reframed to 9:16 with faces kept in the upper third; multi-speaker scenes use split-screen or turn-based centering so nobody important gets cropped.
- Pop or clean captions——Word-by-word pop captions, clean lower-third bars, landscape cuts or no captions at all. Keyword highlights plus per-speaker colors for conversations.
- No cuts mid-sentence——Never cut mid-sentence — every clip keeps one complete idea, padding a second or two where needed. Terms and names are proofread against the transcript before burning.
How it works
- Upload video or paste a link Upload MP4, MOV, WebM or MKV (up to 500MB). File too big? Paste a YouTube, Twitch or cloud-drive link in the box — the agent fetches it when possible. Got a transcript or subtitle file? Attach it and the picks get sharper.
- AI watches and shortlists The agent watches the full video, transcribes it with timestamps, and shortlists highlights by quotable lines, contrarian takes and emotional peaks. You get the list first — timestamps, reasons, opening hooks — and rendering starts only after you confirm.
- Export vertical clips Your confirmed list is cut precisely: no mid-sentence cuts, vertical reframing, burned-in captions, HD polish for fuzzy sources. Each clip downloads as MP4 with a title, hashtags and posting-order advice — lead with the strongest hook.
FAQ
- No video, only photos — can I still make vertical clips?
- No. This tool only cuts long videos into short clips — every frame of the output comes from your original footage, nothing is AI-generated. Turning a photo into a video is a different job (image-to-video): tell the agent what motion you want and it will help, but not through this clipping workflow.
- How many clips can I get from one video?
- It depends on the length and density of your footage. This tool goes for few-but-strong: 1, 3 or 5 finished clips instead of dumping 40 on you. A 60-minute podcast typically yields 3–5 standout moments; a 3-hour stream yields about 5 peak segments. The agent shows you a shortlist with timestamps first, so you decide what gets rendered.
- Do I get automatic captions? Is Chinese accurate?
- Yes. The agent transcribes the full video, then burns the captions in — word-by-word pop captions or clean subtitle bars, with per-speaker colors for conversations. Chinese is supported, but accuracy is best when you upload a transcript file (SRT, VTT, TXT or LRC); tricky names and terms are confirmed against the transcript before burning.
- Can I paste a YouTube link instead of uploading?
- You can paste a YouTube, Twitch or cloud-drive link into the description box. The agent tries to fetch it directly; when a link is login-walled, region-blocked or simply too large, it will say so and ask you to upload the file or split it into parts instead of guessing.
- Is the free version watermarked? Is HD export free?
- Clips exported within the free allowance carry no watermark and are ready to post. If your source footage is low-resolution or heavily compressed, the agent runs an HD enhancement pass before delivery so text stays sharp on phones.
- How long does a one-hour video take?
- The agent must watch the whole video before cutting, so plan on roughly 10–15 minutes of work for an hour of footage — the industry norm for this workflow. You don't wait blindly: the timestamped shortlist arrives first, and rendering starts only after you confirm it.
- AI clipping vs manual editing: what's the catch?
- AI handles the tedious 80%: watching everything, finding the moments, rough-cutting, reframing and captioning. Manual editing still wins for fancy transitions and beat-matched cuts. One honest limit: there is no auto speaker-tracking here — group shots use centered framing, split-screen or turn-based centering instead.
Try it free