1
Describe the scene
Start with who or what is on screen and what happens. One clear action per shot reads better than a list of adjectives.
AI SeedanceText to video
Text to Video AI Generator
Write the shot, get the clip
Describe the subject, the action, the camera, and the sound in plain words. Seedance turns the prompt into a short, directed video — 4 to 15 seconds, 480p up to 4K, with dialogue, ambience, and effects generated in the same pass.
- Clip length
- 4–15 s
- Resolution
- Up to 4K
- Aspect ratios
- 7
- Audio
- Native
This page opens Seedance 2.0 Mini, the lowest-cost tier (20 credits for 5 s at 480p). Switch to Seedance 2.0 for 1080p or 4K in the model selector.
How it works
Text to Video, step by step
2
Add camera and sound
Name the camera move (slow dolly-in, handheld tracking, aerial orbit) and the sound you want; put spoken lines in quotes.
3
Pick length, ratio, and resolution
Choose 4–15 seconds, 16:9 / 9:16 / 1:1 / 4:3 / 3:4 / 21:9 or adaptive, and a resolution. The credit cost updates as you change them.
4
Generate and refine
Review the clip, change one variable at a time, and regenerate. Download the MP4 when it is right.
Why it works
What you can control
Follows camera language
Dolly, pan, orbit, crane, rack focus, handheld — written camera direction is treated as an instruction, not decoration.
Multi-shot in one prompt
Describe two or three beats in order and Seedance cuts between them while keeping the character and setting consistent.
Sound generated with the picture
Turn on audio and dialogue, ambience, and effects are produced with the video and timed to the motion.
Every delivery format
Landscape for YouTube, vertical for Reels and Shorts, square for feeds, 21:9 for cinematic frames.
Best uses
What people make with it
Social clips from an idea
Go from a one-line concept to a vertical clip without shooting or editing.
Storyboards and previs
Test a camera move or a scene rhythm before committing to a shoot.
Ad concepts
Draft several creative directions for the same product line and compare them side by side.
B-roll and atmosphere
Generate establishing shots, textures, and mood footage to fill an edit.
Keep exploring
More Seedance tools
FAQ
Questions about Text to Video
What is text to video AI?
Text to video AI generates a video clip from a written description. You describe the subject, action, camera, and sound; the model renders the motion, lighting, and audio — no footage or editing needed.
Which model does this text to video page use?
It opens Seedance 2.0 Mini by default. Seedance 2.0 (up to 4K) and Seedance 2.0 Fast are in the same model selector, and every tier accepts the same prompt.
How long can a text to video clip be?
Any length from 4 to 15 seconds per generation. Longer durations cost more credits; the form shows the exact amount before you generate.
How much does text to video cost?
Cost is per second of output. Seedance 2.0 Mini is 4 credits per second at 480p, so a 5-second clip is 20 credits — the same as the signup gift for new accounts.
Can I add sound and dialogue?
Yes. Turn on audio and describe it in the prompt: ambience, effects, music mood, and spoken lines in quotes. The audio is generated with the video, not added afterwards.
How do I write a good text to video prompt?
Order it as subject → action → setting → camera → style → sound, keep one main action per shot, and change only one thing between generations so you can see what helped.