Wan2.2 S2V: Music Customization

Introduction

This is a sample native workflow for audio-driven video generation using Wan2.2-S2V, based on ComfyUI.

We are pleased to announce that Wan2.2-S2V, our advanced audio-driven video generation model, is now natively supported in ComfyUI! This powerful AI model transforms static image and audio input into dynamic video content, supporting a variety of creative content needs, such as dialogue, singing, and performance.

Model Highlights

• Audio-Driven Video Generation: Converts static images and audio into synchronized videos
• Cinematic Quality: Generates high-quality videos with natural expressions and movements
• Minute-Level Generation: Supports long-duration video creation
• Multi-Format Support: Applicable to both full-body and half-body characters
• Enhanced Motion Control: Generates motion and environments based on text commands

Recommended machine：Ultra-PRO

Workflow Overview

How to use this workflow

Step 1: Load Audio

Step 2: Load Image

Step 3: Input the Prompt

Step 4: Set FPS

Usually, the default settings are sufficient without requiring multiple modifications.

Step 5: Get Video

You can change the video length by setting frame_rate or num_frames (in WanVideo Empty Embeds). Video length = num_frames/fps