How to key trigger a different mouth (i.e. smile, laugh, frown)

[jonnymackjackson]

Searching for the answers in the forums really isn't efficient because people tend to phrase their questions in vague terms, so even if the answer is out there, it's extremely hard to find and more often than not, it's hard to understand. So when I figure out a way to do a task or fix a problem, I'm going to post a "how to". I learn best with visuals, so the more screenshots, the better. Help out the community and add one or two of your own!

This one was driving me crazy and I couldn't find a clear answer that worked. When recording the "Face", the smile would flash when I didn't want it to and wouldn't show up when I did want it to. So put a trigger on it, right? Not that simple. There's a whole hierarchy of layers and naming items properly (another "How To" for another time) that's hard to understand for dummies like me. What I finally figured out, thanks to trying some things that alank99101739​ described, is that you have to put the Mouth viseme group inside of another group (or layer) in your PSD or AI file and name it something other than "mouth". I named mine "Laugh group".

I put the Mouth layer on top of the Laugh so that I could make a swap set out of the Laugh Group wherein "Mouth" would be the default layer and I could trigger the Laugh with a key (L). Notice I got rid of the Smile viseme--I never want it activated by my face when recording from the webcam--but I imagine you could do both?

I also wanted the Laugh to trigger when he turns his head. Haven't been able to find a way to do it another way, so I just made four more swap sets (after creating four more Laugh groups in each head turn in AI). Seems like you ought to be able to just have one swap set that works for the Laugh triggers in all the views--anyone know if is this possible?

Next I checked to make sure it would work after recording the audio with visemes. Woo hoo! It works!

Screen shot of all the tracks that got recorded from having multiple swap sets. Seems excessive.

Having all the Laugh groups named the same was confusing, so I went back into AI and renamed each head profile's Laugh group:

Bonus tip: just record a couple seconds of the triggered action, then use the Split function to make multiple instances on the same track.

I didn't try it, but I imagine adding other mouth shapes like a frown or grimace would simply go inside the same group.

If anybody out there has an easier way to do this or if has a better idea for how to index answers to basic questions, make it happen! In the meantime, hopefully this helps!

[alank99101739]

Cool! Looks like you got it all worked out with the different facial expressions!

Yes - you can put smile inside Mouth and have it outside in a swapset - but in that case you can just put the trigger on the Smile in the Mouth - the trigger overrides the normal Face behavior and it works nicely.

One trigger can have multiple layers attached to it. So what you can do is if each profile is structured with the same hierarchy, set up the swap set for the first profile profile, then in the rigging hierarchy drag the same layer in each of the profiles over the trigger you created for the first profile. This will add that layer to the existing trigger. That is, the trigger will turn that layer on for all profiles. Since only one profile is visible at a time, you just see one profile. This also means you have have Laugh trigger one while turning the head, it will smoothly be Laugh on for all profiles as the head rotates. I have found this the easiest way, and it avoids clutter in the timeline. If you have different triggers and swapsets, your timeline ends up with a separate take per profile, which gets cluttered. I prefer one trigger bound to multiple layers. (You can also create multiple triggers bound to the same key so all triggers fire at the same time for a keypress.)

When you select a trigger in the trigger panel, it lists all the layers bound to that trigger. Normally its a single layer, but if you drag other layers over the existing trigger name, you will see a list of layers like

RightProfile/Face/Mouth/Laugh

LeftProfile/Face/Mouth/Laugh

etc

There is a little x next to each layer in a trigger, allowing you to delete layers from a trigger as well.

[jonnymackjackson]

First time I'm having the chance to try this out and I've tried seemingly everything--can't figure out what I'm doing wrong.

First I tried just adding the Laugh layer from the other profiles into the Frontal swap set--the trigger works, but the Neutral mouth stays there (in the other profiles--the Frontal profile does replace the Neutral).

So I tried renaming the LQ Laugh group to match the Frontal--so they both were Frontal/Laugh Group/Mouth/Laugh. Same problem--trigger works, but both mouths show.

Then I tried removing the Laugh Group and just put the Laugh inside the Mouth group, so it's Frontal/Mouth/Laugh. Now the trigger doesn't work at all. Tried putting Laugh at the bottom of the Mouth group, then put it directly under Neutral, then as the top layer of the Mouth group---same--trigger doesn't work at all.

So far the only thing that's worked is the way I had it--with different swap sets for each profile, but like you say, it creates a mess of layers.

Any ideas?

[jonnymackjackson]

Of course I figured it out right after pressing send.

In addition to putting each Laugh inside the Laugh trigger, you also have to put each Mouth (default) into the swap set. I assumed it would just know that I meant for the default layer of each group to disapper when I put each Laugh into the swap set----I guess computers aren't as smart as we think they are---yet.