🎬 Generation Modalities

📝➡️🎥 TextToVideo
Create completely new videos from scratch using text prompts or optional audio input.

🖼️➡️🎥 ImageToVideo
Animate static reference images using text prompts or optional audio input.

🎥⚪ VideoToVisualDub
Generates synchronized audio tracks such as ambience and speech, driven by the video visuals and prompts, while keeping the original video.

🎥⚪ VideoToMaskedFaceGen (Warning: With the new ComfyUI update v0.9.1, Inpaint doesn’t work anymore.)
Regenerate masked facial areas. Control expressions, lip-sync, and identity using prompts or optional audio input.

ℹ️ Info: The input video resolution after scaling is the same as the output video resolution. Internal spatial downscaling or upscaling is deactivated.

🎚️ Audio Input Settings

⚪ 🔇 No Audio Input
No external audio file is used. The AI generates completely new audio based on your text prompt.

⚪ 🔊 ++ Audio Input
Upload an existing voice or music file to drive the animation, for example for lip-sync.

* Note: Do not use this setting for VisualDub.