Input Requirements for Optimal Results in Image, Video, and Audio Generation

[WeizhenEricFang]

Hello,

I’ve tried uploading several images and videos, but the results haven’t been satisfactory. Could you please clarify the input requirements for generating good results? Specifically:

What are the ideal conditions for the images, videos, and audio inputs?
Are there any specific recommendations regarding aspect ratio for inputs to ensure high-quality outputs?
I would appreciate any guidelines or best practices that could help improve the results.

Thank you!

[digital-avatar]

@WeizhenEricFang
Can you explain the specific circumstances of poor results?

Generally speaking, the most important requirement is that the source image mouth is closed, and the audio requirements are not too strict. If the pronunciation is clear and there is no other sound except speaking, the driving effect will be better.