Lip-Sync - How do you all do it?

[Eddy]

Lip-syncing in animation has proven to be one of the most boring, mundane, and tedious tasks for me, but I just can't stand it when the mouths don't match what they're saying. So, basically, my question is, which program(s) do you use to sync up your characters to what they're saying?

I've been using Papagayo for pretty much all of my short animation career so far, but, helpful as it is, it's far from perfect. The pattern it produces never matches up with the voices well, so I end up spending a good amount of time (actually, most of the production time for me is this) fixing it up so that it actually syncs. Then, I count off the frames and lip-sync it to Flash, since it doesn't have an export.

I can live without the export, but what I really, really would like is a program that can produce syncing that doesn't need me to spend half an hour tweaking a five second lip sync pattern.

So, I thought I'd ask you guys - what do you all use, and what would you recommend to others?

[Vincent Veidt]

Now that I've read everyones' responses, I can freely input. Toon Boom Digital Pro (as well as all their animation products) has a decent system that's based on phoneme templates you create and an X-Sheet breakdown (anyone who's converted dialogue to phonemes in an x-sheet knows the trick, if you don't, refer to Richard Williams Animator's Survival Kit).

Now that I've finished that, one of the few openly agreeable things John K. has brought to the table is that all lip sync should be done by hand. Forget about templates, they suck the emotion and life from your character. Just do it the old fashioned way for best results.

3) The general consensus about lip sync timing is two frames ahead. I've heard of people pulling off more and less, but two frames (at 24/s) works pretty darn well.

7) Don't pay too much attention to details. That's what's really killing me now. If you concentrate too much effort of minor details that will be invisible in their 12th of a second on screen (assuming you animate on twos), you will miss deadlines and be difficult to employ... unless your emloyer is into that (and is willing to wait a year for two minutes of animation).

5) Good call, Stephen, a computer is just a tool. Nothing can replace talent, practice and willpower. Just keep going and try not to lose yourself in details (like the 900 notches I put in every shoe... just kidding, I've cut it down to 7).

[Damian Byrne]

Evening!
Some good points here.
Ive a few that may interest you. Im currently working on a 2d film in Uni and have had to come up with some speedy ways to work since ive taken on such a big project anyways...
If you rememeber light travels faster than sound then your on the right path. ie, we see before we hear (in this example).
Possibly place the key 3-4 frames ahead of the sound. I wouldnt bother with Simpsons to be honest but rather Anime and in their language. You have to rememeber your representing sound with your pictures. Its the phonetics were looking for here (aaaaw, eeeew, nnnuh etc) . So if you draw the extremes (depends on the character, human= not so much but actually very minimal.
USE A MIRROR!!
Monsters= you can go nuts! or vice versa) of the sound then in my case import them into Adobe Premier. If you can have the KEYS in photoshop (individual layers) they will tansfer to Premier. Import the audio file and then you can see the wave file. This is very helpful as you can use both your eyes and ears to sync up your keys.
Hope that helps.
Best.
D.
Post some up online so i can have a better look?

[Mike Leins]

I've never done lip syncing myself, so it's good to see all the responses here. I always wondered how a computer program could guess what is being said anyway. Transcription software tends to do a crappy job of turing audio into text, how good could the computer do at animating based on an mp3 anyway?

In relation to lip syncing, I've heard that, although it looks like there are tons of mouth movements in most cartoons, the reaility is that there are only 10-15 mouths that need to be created. All vowels, along with a few consonant sounds are needed, and some can double up, such as "sh" and "ch" aren't noticeably differnt, so why draw two separate mouths?

[Damian Byrne]

Hey hey!
If i recall way back when i learned Max i copied the head many times and then changed the mouth shapes to represent the 'phonetics' of a sentence on each head. Head 'EE', head 'OWW' etc. Actually i exagerated them so that when they were 'imported' (not sure what the technical term is for your software) i had more options to play around with. Im afraid Lip syncing is an art in itself. However, its true though, that you can get away with alot with very few mouth shapes as mentioned earlier in ANIME. As a beginner id go for the most extreme sounding and then use them as a kind LIMIT. If that makes sence. This is really only as a gauge so then you can have an averall feel for the line before you polish it up with inbetweens, half mouths, halfs 'ww's, etc etc.
Good luck.
I hope to post some 2D lip syncing soon!!

[Heather McColgan]

I have anime studio, the programme does lip syncing its self, you just import the files into the programme. Theres usually 5 mouths that you have to make for it to work. Anime studio uses open, closed, fully opened mouth etc.