Hello,
This is not meant as a joke or snark in anyway what so ever.
This all came about due to not getting good results from this app after months of training as model. ( Literally, 3 months of cumulative 24 hour days.)
While I bounce around from various AI related projects, I have encountered a special scenario with stable diffusion. Long story short, no matter what I did there, I could not train a custom model of a face using Dreambooth. The face was always "off". Seems to be a theme for me 
I found and enabled an extension in Stable diffusion called FaceSwapLab. With it, you can generate an image and then swap the face. You can also swap the face of a previously generated image, and it does a really good job of matching the expressions. Not perfect, but still, pretty good.
So, I am working on a script where I generate my swap images using that extension and will then need to figure out how to do the conversion. ( I have hopes that it won't be so hard)
My question is has the author of this program considered doing something similar or even using the existing extension I mentioned? I do understand that they are probably competitors, but it would save SOOOOOOO much time and effort.
Thanks for reading and for the awesome(ish) product.
