Make AI Videos Using Clips You Already Have

People use the words "AI video" for two different things. One kind of tool generates new footage from a text prompt. The other kind takes real footage and edits it for you. This page is about the second kind. When you make AI videos using clips here, every shot in the picture comes from a file you uploaded.

The AI does the editor's work, not the camera's work. It watches each clip, reads your script, matches a shot to each line, trims it to the right length and cuts it on the beat. Then it adds a voiceover and captions. The result is an AI video from your clips that looks edited by hand. If you want new footage from a prompt instead, use the AI video generator.

Many people search for a way to make AI videos using YouTube clips. This tool does not download anything from YouTube. It edits clips that you upload. So the question is where your clips come from, and whether you have the right to use them.

Good sources are footage from your own YouTube channel (you can download your own uploads from YouTube Studio), your own shoots, and stock or archive clips you hold a license for. Reusing other creators' videos needs their permission, and it can trigger copyright claims on your channel. With clips you own or license, you can make an AI video from clips without that risk.

First, create a project and pick 16:9 for YouTube or 9:16 for TikTok, Reels and Shorts. Drop in every clip you might use. Your browser compresses heavy footage, such as 4K iPhone clips, to a light 1080p copy and uploads it, with the progress of each file on screen.

Second, paste the script. The editor splits it into caption-sized phrases. Third, start the match. AI reads every clip in batches of 8, writes a short description of each part, and then ranks the parts that fit each phrase best. A fixed assembler cuts the shots at the pace you chose.

Last, read the timeline. When it reads well, render one MP4 and download it. To make AI videos with your clips again later, keep the project: clips stay until you delete it, and the finished video is also saved in My Works.

An AI video editor can only show what your footage holds. Before you upload, read your script and think about what each line needs to show. Upload clips for every idea the script mentions, and several options for the ideas that come up often.

Short, steady clips with one clear subject each are the easiest to describe and to match. Each clip can be an .mp4, .mov or .m4v file of any size, or a photo in .jpg, .png, .webp or .heic, and there is no limit on the number of clips in a project. For very large libraries we recommend the free desktop app for Mac and Windows, because the web editor gets slow with very many or very large files. If an iPhone .mov file fails, it is usually HEVC: try Safari, or export the clip as H.264 MP4.

The script is the plan of the whole video. Each phrase becomes one caption and one beat of footage, so short sentences give the clearest cuts. Write in the order you want the viewer to see things, and the AI video editor follows that order.

Concrete words give the best matches when you make AI videos using clips. A line such as "We packed the boxes at dawn" points at a picture, while "It was a big day" could match almost anything. When no clip fits a phrase, the least used clip stands in as a filler. The timeline marks it, so you can add footage and match again.

The cut follows the timing of a voice. An AI voice reads your script in English, or you upload your own recording. Every caption and every shot is timed to the moment the voice says the words. Add music from the presets or your own file, and choose a caption style.

The pace decides how often the picture changes. Fast gives shots of about 1.5 seconds, which suits a Short. Slow gives each shot up to about 4 seconds, which suits a calm explainer. Pick 16:9 (1920 x 1080) or 9:16 (1080 x 1920). Long videos render in parts of about a minute, so a change renders only the parts it touched.

A faceless YouTube channel is a natural fit. Write the narration, upload the footage for that episode, and make an AI video from clips instead of building the timeline by hand. For a full YouTube workflow, see make a YouTube video with clips.

You can also turn your own long videos into Shorts. Upload the parts you want, write a short script, and pick 9:16 with the fast pace. Teams with a folder of old shoots can build a new explainer from archive footage. Product teams can match their own product clips to a demo script. When you only need to merge clips in the order of a story, see combine video clips.

Your first AI video from clips is free: up to 60 seconds of script, 50 clips and one AI voice or your own recording. After that, each step uses purchased credits. Reading clips costs 1 credit per batch of 8 clips, matching is priced per part of the script, and rendering costs 2 credits per part of about a minute, plus 1 credit to join the parts.

To spend less, upload only clips you might use, and settle the script and the pace before you match, because a change to either one matches again. A failed step is refunded and the finished work stays, so a retry pays only for what is left. For every setting of the editor, open the AI B-roll editor.