Use Seedance 2.0 to create lip sync videos from text prompts, reference images, short reference videos, and audio. Write the exact words your character should say, or upload reference media and audio so Seedance can follow the speech.
The #1 video generation model for ads, films, and product creative. Supports text, image, video, and audio references, and you can target reference assets in prompts with @image1, @video1, and similar tags.
Reference Images (Optional)
Upload up to 9 reference images in a single area.
Total reference video duration must not exceed 15 seconds.
Total reference audio duration must not exceed 15 seconds.
Log in to get daily credits and start generating videos. Your tasks will continue in the background if you close the page. Please do not submit the same task repeatedly. You can find your previous generations on the My Creations page.
Reference Images (Optional)
Upload up to 9 reference images in a single area.
Total reference video duration must not exceed 15 seconds.
Total reference audio duration must not exceed 15 seconds.
Log in to get daily credits and start generating videos. Your tasks will continue in the background if you close the page. Please do not submit the same task repeatedly. You can find your previous generations on the My Creations page.
Generated Videos








































Explore Seedance videos made with prompt dialogue, image references, video references, and audio-guided lip sync.
A panda sits on the left and looks at the camera, saying, "Hello everyone." After that, a raccoon on the right speaks and says, "Welcome to Lip Sync Studio"

Use the song from @audio1 to generate a video of a man singing.
Standard
* Annual credits are issued in full upon purchase and refreshed annually.
Pro
* Annual credits are issued in full upon purchase and refreshed annually.
Basic
* Annual credits are issued in full upon purchase and refreshed annually.
Pay as you go. Credits never expire.
Seedance can create lip sync in two practical ways: write the spoken dialogue directly in the prompt, or upload reference video plus reference audio and instruct the character to speak the audio content.
Describe the character, scene, emotion, and exact line of dialogue. Put the words in quotes and add phrases such as natural lip movement, clear articulation, direct eye contact, and subtle head motion.
Upload a reference video or image for the character and an audio reference for the speech. In the prompt, tell Seedance to follow the uploaded audio for lip sync and preserve the character, motion style, and camera framing from the reference media.
Seedance 2.0 supports text, image, video, and audio references. For lip sync, keep prompts specific: identify the speaker, quote the dialogue, and reference uploaded media with clear instructions such as use @audio1 for the voice and lip movement.
Upload character images, a short reference video, or clean speech audio when you need tighter control.
Include the exact words the character should say, the speaking style, and any @image, @video, or @audio references.
Choose duration, aspect ratio, resolution, and whether Seedance should generate audio in the same pass.
Preview the mouth movement, regenerate variations if needed, then download the final lip sync video.
Start with a short 5 to 8 second clip. Use a clean prompt, clear spoken words, and a close-up character reference for the best first result.