The facial animation is still not symmetrical in audio2face

The new version of Audio2Face claims there’s no a support for symmetrical blendshape export. I have setup the linking of particular blendshapes. However the result is still not symmetrical in the end when I export to Unreal Engine and play it on my metahuman.

Is there a compreshensive guide on using symmetry?

Also it would help to have a genereated face template that is perfectly symmetrical.

Our aim is not symmetry unfortunately. A lot of realism comes from asymmetry in how our face moves. But we understand there are many use cases that need / prefer symmetrical output.

What are you thinking about using this for that you like to be symmetrical?

I want to make sure that Symmetry Pose is working. This was the latest feature added recently.