MoDA: MODA PLUS: Talking Head Generation

Enhanced Version with Smooth Motion

📥 Input Settings

Source Image

Driving Audio (Recommended: < 60 seconds)

0:00

⚙️ Generation Settings

Emotion

Select an emotion for more natural facial expressions

CFG Scale (Lower = Smoother motion)

Lower values produce smoother but less controlled motion

0.5 5

🎬 Motion Enhancement

May cause errors on some systems. If errors occur, disable this option.

Enable Smoothing (Experimental)

Target FPS

Higher FPS for smoother motion. 30 FPS recommended, 50 FPS maximum

24 50

📺 Output

Generated Video

Tips for best results:
• Use high-quality front-facing images
• Clear audio without background noise
• Keep audio under 60 seconds
• Adjust CFG scale if motion seems stiff
• For longer audio, split into segments

Example Configurations

Source Image	Driving Audio (Recommended: < 60 seconds)	Emotion	CFG Scale (Lower = Smoother motion)	Enable Smoothing (Experimental)	Target FPS