Should Animation Montages Be Used for Facial Metahuman Animations?

Is it common practice to use Animation Montage for facial metahuman animations to create phrases? Or is it better to stick with a State Machine for this?

If Montages are the way to go, why is there no tutorial on how to create an Anim Montage for the face, how to add a slot to the face animation blueprint, and make it all work? I tried creating my own slot and added it to the face animation blueprint, but the animations stopped working after that. Using the Default slot works fine, but the body and head blending becomes incorrect.

Or should I just not overthink it and continue using the State Machine like I’ve been doing before?

Well state machines work well with locomotion so would be better using a layered blend per bone node if using facial clusters. If using morph targets you don’t need to use layering as morph targets are additive when added to the mesh level.

Montages are difficult to work with when sharing the same slot as triggering the shape will override the current montage.

Are you using morphing or clusters?