AI long video maker
Give it a topic — or your own script — and get a narrated video up to five minutes long: a voice-over, a picture for every scene, music underneath and captions on top. You read the script before anything else is made.
1 · What is it about
2 · Look and sound
What it will cost
- Script2
- Images — 618
- Voice-over5
- Music bed10
- Assembling the video0
- About35 credits
You pay for the script first and see it before anything else is made. Anything already made is reused — a retry or a reload only pays for what is missing.
Script first, then the video.
The script appears here scene by scene. Read it, change what you like, then make the video — narration, one picture per scene, music and captions, up to five minutes.
How it works
- 1. The script. A title, then one scene for roughly every 10 seconds: the words the narrator says and a description of the picture shown while they are said. It is written a minute at a time, so it arrives in seconds rather than after a long wait.
- 2. Your edit. Change any sentence, rewrite any image, remove a scene. Nothing but the script has been paid for yet.
- 3. The parts. The voice-over, one image per scene and the music bed are made in parallel by the same engines as the individual tools, and saved to your gallery.
- 4. The video. Your browser puts it together — camera moves, cross-fades, captions, the music ducked under the voice — and hands you an MP4 and a subtitle file.
What it costs
There is no separate price for this tool. A long video costs exactly what its parts cost: 3 credits per image, 5 credits per 2,000 characters of narration, 10 credits for the music bed, and 2 credits per minute of script. Assembling the file is free, and so is assembling it again.
| Length | Scenes | Credits, with voice and music |
|---|---|---|
| 1 minute | 6 | about 35 |
| 3 minutes | 18 | about 80 |
| 5 minutes | 30 | about 125 |
Animated scenes add 12 credits each, and a Multilingual narrator costs more per character. Both show their price before you start. See the full price list.
Why the video is assembled in your browser
Five minutes of video is more than seven thousand frames to encode. On a server that means a queue, a timeout and a cost that ends up in the price. Your browser has a hardware video encoder built in, so the file is written on your own device, usually faster than the video plays, and the assembly costs no credits — the first time or the tenth.
What it will not do
- • It does not generate five minutes of moving footage. Scenes are stills with camera moves; up to five can be animated into short clips.
- • Scene changes are timed from the length of each scene's text, not from the audio itself, so a cut can land up to a second early or late. Under a narrator that is invisible; for lip-sync it would not be.
- • One narrator per video, one music bed, one caption style. For anything finer, the parts are in your gallery and the free editor takes them from there.
- • It needs a recent browser: Chrome or Edge, Firefox 130 or later, or Safari 26 or later. Where the browser has no MP4 encoder the file is WebM, which YouTube and most platforms accept as it is.
Questions people ask
How long can the video be?
Up to five minutes when the tool writes the script, and a little over six when you paste your own (6,000 characters of narration). Lengths on offer are 30 seconds, 1, 2, 3 and 5 minutes.
What does a five-minute video cost?
About 125 credits with a standard narrator and a music bed: the script, 30 images, the voice-over and the music. Assembling the video is free. The exact figure is shown before you start, and it is simply the sum of what the image, voice and music tools charge on their own.
Is it real video, or a slideshow?
By default each scene is a still image with a slow camera move, cross-faded into the next under the narration — the format most explainer and documentary channels use. You can have up to five scenes animated into moving clips for 12 credits each. A fully generated five-minute video, every second of it, is not something any engine does well yet, and we would rather say so than sell it.
What happens if something fails halfway?
Nothing is lost and nothing is charged twice. A part that fails is refunded on the spot; everything already made is kept, and pressing the button again makes only what is missing. That holds if you close the tab or lose your connection too.
Can I use my own script?
Yes. Paste it and it is cut into scenes at sentence ends; the tool only chooses an image for each one. Every word and every image description stays editable before the video is made.
Which languages can it narrate?
The standard narrators read English and French. The Multilingual narrators read 29 languages, including Spanish, German, Arabic and Hindi, at a higher price per character that is shown before you start.
Where is the video stored?
The images, the voice-over and the music are saved to your gallery. The finished video file is assembled in your browser and downloaded from there — it is not uploaded to us, so download it to keep it.