Thursday, March 26, 2009

Do you Have a Set Way on How You Go about Lip Synching? Is There Any Method You Have Found the Most Effective?

Lip synch is one of my favourite parts of animation, along with hand animation. I find that those two elements really 'plus' the expression and finesse great animation. There are some shows that the lip synch is my favourite part…like Southpark!

Anyway, there are several things I can share about how I go about lip synching that might be helpful to other animators out there. One is, I have a trick that I use to help me work out how much the jaw is moving. That's the first thing I tackle when I'm starting to block out the dialogue, and I find that the hardest thing to do is to NOT make the jaw move *too much*. The jaw actually moves a LOT less than one might think - which can be particularly deceiving when you're dealing with fast-talkin' dialogue.

What I do is I hold the back of my hand up underneath the bottom of my chin, and repeat the dialogue that I'm blocking out. The trick is to repeat it out loud, with the exact same inflections, volume levels, and even the same accents as the voice actor. Yes, people might look at you funny, but we must suffer for our art! :)

So as I'm repeating the dialogue (the more I repeat it, the less I have to think about what word or inflection is coming up, so I can concentrate on the movement), I start to pay attention to how often, when, and how much my jaw opens up. It is this movement that I use for reference for blocking the jaw action initially, and I find it creates a great base for the lip synch animation. It's much the same as blocking in the hip action for body animation... it's that base that can make or break the rest of the action that is layered on top of it.

Something else that I find is true for me is that once I've got all of the jaw, lip and tongue movement working (mostly polished), I always grab ALL of the keys and shift them down the timeline by two frames. I find that when I get the lip synch working with the maya file, and I can hear and see exactly when the sounds are formed on the timeline when you scrub through it, it *always* appears to be a touch early when I playblast and watch it back at full speed. What crazy phenomenon causes this? Could it be that the speed of sound is slower than the speed of light, and so I'm hearing the sound ever-so-slightly after I'm seeing the lip shapes formed?

Well, whatever the reason, when I shift the keys two frames later, it appears correct. I've found that most people have a pattern like this in their lip synch animation, but not everyone's is the same. For example, one animator might only have to shift it one frame, or even one or two frames *earlier* instead of later. But if you find you're having trouble with getting your lip synch to synch... then this might be the answer for you!

Guest Blogger Dana Boadway

4 comments:

Herman said...

just in time, 1st pass lip sync 4 me right now. I never heard of that tip of the hand under the jaw, very cool. thank you very much.

Now I curious with the fingers and hand now since you mentioned it, just handle it like shape changes?

3th Dimension said...

Thx a lot dana for your advices!!
Always usefull.

parallelogramtastic said...

Thanks for the tips! I find I have a similar method for digital animation: (1)ups-and-downs, (2)ins-and-outs, (3) tongue and polish.
(1)The ups-and downs is pretty much the same idea with figuring out the jaw movement. I also try to figure out when the lips will be open and closed on this pass.
(2) The ins-and-outs pass is for figuring out all the "ees" and "oohs" for the dialogue and adjusting the phonemes accordingly.
(3) Tongue and polish. I always used to have problems adjusting my animation and then finding that the tongue now poked through the geometry, so I save this until last, along with polishing where I will exaggerate the expression and multiply blend shapes on top of the dialogue...

Virgil said...

I've heard that speed-of-light vs sound thing many times and I think it makes no sense, as both speeds are enormously huge, and the difference between them is clearly significantly less than 1 frame, or 1 24th of a second. I hoped someone might have discovered the real issue here, but I guess it's still a mystery.

My (undocumented) theory is that the sound is formed in the throat/mouth a bit earlier than we actually release it visually by opening the mouth (or something like that...) So... I think vowels should start much earlier, and explosive consonants like P, B, should be more 'on the money'.