While making an original character speak, I have tried creating a separate pose for each phoneme, but find that in an action the frames between one mouth pose to the next are unmanageable. My solution so far has been to create poses for phrases, each phoneme pose set at, say, increments of 5% in a percentage pose. Is there a less tedious approach that allows you to use a set of phoneme poses and animate between them smoothly?

I presume you have done the lipsynch tut in TAoA:M and are doing that procedure correctly. If not give it a try.

However, animators have been getting away from preset mouth poses for lipsynch because of the same problems you have encountered and tend now to directly keyframe the mouth shape for speech.

At a minimum this will use a control for wide-narrow on the mouth and another for jaw up-down and maybe a third for puckering-curling the lips. More elaborate schemes are used but just that will get you decent results.

Jason Osipa's book "Stop Staring" deals with this approach.


But it's not less tedious. It's a bit more work for better results.


Actually you can do a lot with just jaw up-down is your body language is doing its part. Here's one I did for an assignment where we were only allowed to use jaw up-down.

Thanks Robcat -

Man, what you've done with body language and a great character voice is amazing! It really does work with simple mouth movements! I'm not at the level of animating the caricatured body that I need to be but am working on it. The link to that video helps, if only to see how well it can be accomplished in the right hands with limited mouth movement.

Thanks for the input and advice. I have Jason Osipia's book but haven't looked at it in awhile. The character he uses that graces the cover is classic and I'm assuming the lighting effects on the face (relative transparancy around the ears, etc.) are accomplished with subsurface scattering. That's another topic I haven't learned yet.

Back to it!

Thanks. Yes, everything is better when Orson Welles is doing the talking.



I have the first edition which included some coverage of A:M in it although he seemed to be on an ancient version.

LOL! Yeah, Orson Welles . . . I didn't place his voice though it did sound familiar . . . not a good sign since I just watched John Huston's adaptation of "Moby Dick" with Welles as the preacher!

I've been looking at lip-synch myself - but I find the tutorial in the booklet doesn't work with my own model and my own test sound-clip.
I use those "Dope Sheets", but I find that even after setting each word to its correct place on the soundtrack, the mouth still moves out of synch with the sound.
I'm thinking I'll have to do this manually - like I do with Flash, which I incidentally find that a lot better for simple 2D animation!

The trouble is, most people do not realise that the mouth shapes have to preceed the actual sound. The most common mistake with the dope sheet is lining up the word with the sound of the word - the lips are going to be late this way. The other thing is that if you do the dialogue as a separate action then bring that into a chor then the dialogue is goiing to be generally late by 1-2 frames. So in the action, when you are stisfied with your results, pull all the keys left 2 frames then refresh the chor.

Cheers