Does the model support style-preserving text replacement (Scene Text Editing)?

[zqpresent]

Hello,

I have a quick question about the model's capabilities.

Does this model only support synthesizing text in a specified region (text generation/inpainting), or does it also support style-preserving text replacement?

Specifically, I am looking to see if the model can use the original text in the masked area as a style reference to generate new text that matches the original font and look (similar to Scene Text Editing tasks).

Thanks!