Text to Video

Write freely, or use "Scene 1:", "Scene 2:" and so on. Scenes are split by those markers, then by paragraphs, then by sentences. Each scene's description picks its visuals; the caption is what shows on screen. Editing this box rebuilds the scene cards.
Each upload goes to the scene whose description best matches its file name or tag. Drag an upload onto a scene card, or pick it on the card, to place it yourself. Stock media only fills scenes without an upload.
AI images are made for free on our own server. Each takes about 10 to 15 seconds.

Voice uses your browser's speech and is for preview only. It will most likely not be included in the exported video.

The Read narration box also adds a stock computer voice to Real video exports.

Do not request or upload content depicting real people's likeness, or copyrighted or trademarked material, unless you own the rights. The user assumes all liability for prompts, uploads and generated videos.
Real video builds each scene from real, freely licensed video clips (Wikimedia Commons) that match your words, plus motion, crossfades and an optional narrator. It is matched stock footage, not AI-made video. If it is unavailable, use Generate and edit.

All rendering, editing and exporting happens in your browser; nothing is uploaded. The generated video opens in the editor below, where every scene stays editable.

Want more AI tools? See the AI offer.