lip synching made EASY(er) - a quick tutorial

The old Video Software Help forum, left visible as an archive.

lip synching made EASY(er) - a quick tutorial

Postby iamfanboy » Sun Oct 22, 2006 6:33 pm

I was reading the tutorial available on the website about how to do lip synching, and it misses a VERY important notation on the theory of lip synching.

(helped me out a lot, though; I didn't even know there was a zoom function in Premiere 'til I read it ^^; )

Searchy, also, fails to find a tuturial on the forum dedicated to this. There's a casual notation in one of the threads, but nothing more than that - and if someone needs help, will they find it there? Better to have an emblazoned thread that says, "FIND IT HERE!"

Lip synching by itself (not counting fancy pans or other movement in the shot) is not hard, animators have been doing it for almost eighty years now. Why can't you?

No, it's not because you're dumb. I'm scandalized that you would even think such a thing! It's because they knew something you didn't: the basics of lip movement.

While I was tempted to go and pull some frames from an old Disney cartoon for this, instead I decided to use what I had and do this quick, while it was still in my head.

Essentially, the human mouth has 'three' positions when talking, and these are all the frames that YOU need when you're doing lip synching. You can have more than this, and your work will look better, but it'll be time-consuming and not altogether necessary. The eye will be tricked into 'seeing' the words with just these three mouth positions.

The MOST (MOST MOST MOST) common is halfway open:
Image
Usually if a tongue movement makes a noise (like 'ch', 'ka', etcetera) then it's a halfway-open moment.

Next most common is fully open:
Image
For yelling, of course; but also for sounds like 'o', 'ha', 'e'.... a lot of vowels by themselves fall in to this.

And LEAST common (pay attention, now!) is fully closed:
Image
Only a few, sharp sounds use this, like 'p', 'th', and others.

I'm sure anime studios have little "Basic Mouth Movement" charts up on their walls, with each of the Japanese hiragana identified with a mouth size so that amateurs don't get it mixed up. Non necessary for us, because we've got their work right in front of us.

But how to do it easily? Nothing will replace effort and patience, but there are a few things to understand that will make it easier.

The halfway-open position is the most common for mouths to be in while talking. I can't stress this enough. Humans almost NEVER close their mouths while talking except to make very specific noises. If there's no other movement necessary in the shot, I'd suggest exporting the halfway-open frame as a bitmap and placing it on the bottom video layer (Video 1B, for Premiere users), so you don't waste a lot of effort copying the most common mouth position. It looks static, but if that doesn't matter it will make it easier until you can get the hang of this thing.

The closed position is the least common for mouths to be in while talking. Why am I repeating this? Because it's such a simple mistake to make, but it's so common and so unnecessary. Don't even have a closed mouth BETWEEN words; do you really close your mouth between each word? I'm serious. Leave it in the halfway position between words. S'all right to have it closed after they're done speaking, though; I shall allow you that much. :P

Never do single-frame flashes of a mouth movement. You want at least 2, more commonly 3-4 frames, for any mouth movements. The only exception should be SOME of the closed-mouth sounds, which are always very fast, but you should always have a halfway-open frame before and after the close mouth.

This gets said over and over, but seriously, get a mirror and SEE how your mouth moves when you speak whatever you want to lip synch. Watch each word carefully and divide it up into halfway, fully, and closed portions. If you don't and you still complain, you have no one but yourself to blame.


And now for the bit of a pimp (small price to pay for a handy tuturial?) I managed to do some reasonably effective lip synching in this video, the first I ever tried lip synching in, because I understood these techniques, and now hopefully, you do too.
http://www.animemusicvideos.org/members ... p?v=131422

The other tuturial at http://www.animemusicvideos.org/guides/ ... psync.html is fine (great!) for the technical aspect; if you don't know how to do it or even what I'm talking about when I say 'frame' (don't laugh, I didn't know either) go read it right away. But as far as it goes he didn't EMPHASIZE the theory of it enough, or even have more than a single sentence on it - and he missed the most important point, that a closed mouth is least common. ^^;

Any other thoughts on the mystic art of lip synching, as long as I've got this thread here ready to recieve your confessions?
Official Fanboy Disclaimer: The above post is not meant to be taken internally. If you do so, consult a doctor immediately. Examine the post carefully, because it may have been meant as tongue-in-cheek. If you take the post too seriously, be warned that the poster may be laughing at you very loudly. Do not taunt Happy Fun Post.
User avatar
iamfanboy
 
Joined: 16 Oct 2006
Location: your pocketses *gollum*

Postby AMV_4000 » Sun Oct 22, 2006 6:37 pm

this should be in software help, or atleast somewhere in the help section..


as for the topic, i usually just take 3 frames, put them on 3 layers, and fade between them... much like you say.. but lipsync has gotten so easy.. what you need to do is a guide on motion lipsyncing...
User avatar
AMV_4000
 
Joined: 10 Apr 2002
Location: USA

Postby Gepetto » Sun Oct 22, 2006 6:52 pm

AMV_4000 wrote:what you need to do is a guide on motion lipsyncing...


He'd have to cover a lot more than just syncing to do that (masking, paths and the like). But it would be a hand on the wheel for the new editors.
And God spoke unto the Chicken, and He said: "Thou shalt crosseth the road", and the Chicken did cross the road, and there was much rejoicing.
My DeviantART profile
User avatar
Gepetto
Mr. Poopy Pants
 
Joined: 10 Jun 2004
Location: The Tokyo Settlement
Status: Bored to tears

Re: lip synching made EASY(er) - a quick tutorial

Postby Scintilla » Sun Oct 22, 2006 6:52 pm

iamfanboy wrote:Never do single-frame flashes of a mouth movement. You want at least 2, more commonly 3-4 frames, for any mouth movements. The only exception should be SOME of the closed-mouth sounds, which are always very fast...

I've been saying this for years. Don't know how much good it's done.

Also, if you're working at 29.97 fps rather than 24 or 23.976, you'll want to do more 3-frame shots and fewer 2-frame ones.

Also, you usually want to make sure that the rest of the character's head is almost or completely stationary between the three reference frames -- it usually looks odd to see parts besides the mouth bobbing back and forth. The one exception I've seen so far is <a href="http://www.animemusicvideos.org/members/members_videoinfo.php?v=117913">this video</a>, which manages to break this rule and still look good doing it (I'm thinking specifically of "the King of Spain never rushes").
ImageImage
:pizza: :pizza: Image :pizza: :pizza:
User avatar
Scintilla
(for EXTREME)
 
Joined: 31 Mar 2003
Location: New Jersey
Status: Quo

Postby Moonie » Sun Oct 22, 2006 7:03 pm

Nice!
This might be Sticky worthy ^_^
ImageImageImage
ImageImageImage
User avatar
Moonie
Give me some candy!
 
Joined: 17 Aug 2003
Location: Garland, Tx
Status: HP:11900 Strength: 940 Gold:60000 MP:1200

Postby Gepetto » Sun Oct 22, 2006 7:07 pm

moonie211=mod? o.o

Better yet, it might be thrown in the GUIDES section if it's given the right formatting.
And God spoke unto the Chicken, and He said: "Thou shalt crosseth the road", and the Chicken did cross the road, and there was much rejoicing.
My DeviantART profile
User avatar
Gepetto
Mr. Poopy Pants
 
Joined: 10 Jun 2004
Location: The Tokyo Settlement
Status: Bored to tears

Postby Melanchthon » Sun Oct 22, 2006 7:09 pm

iamfanboy wrote:You want at least 2, more commonly 3-4 frames, for any mouth movements. The only exception should be SOME of the closed-mouth sounds, which are always very fast

The closed-mouth frames have the same fps as the open / half-open frames, so it'll still be two or three (assuming you're working in 24fps or close) frames at least. I've yet to see otherwise.

The rule of thumb, I think, is to follow the original framerate. Even very small mouth movements look odd if they're synched at 24fps.
User avatar
Melanchthon
 
Joined: 02 Sep 2004

Postby Moonie » Sun Oct 22, 2006 7:19 pm

Gepetto wrote:moonie211=mod? o.o



Id probally make a good mod ^_^
Lets see...


[Moonie211: Im moving this thread to the General Video section and making it a sticky.]


Did that work?
ImageImageImage
ImageImageImage
User avatar
Moonie
Give me some candy!
 
Joined: 17 Aug 2003
Location: Garland, Tx
Status: HP:11900 Strength: 940 Gold:60000 MP:1200

Postby Gepetto » Sun Oct 22, 2006 7:37 pm

Fail. Back to mod school.
And God spoke unto the Chicken, and He said: "Thou shalt crosseth the road", and the Chicken did cross the road, and there was much rejoicing.
My DeviantART profile
User avatar
Gepetto
Mr. Poopy Pants
 
Joined: 10 Jun 2004
Location: The Tokyo Settlement
Status: Bored to tears

Postby Moonie » Sun Oct 22, 2006 7:49 pm

Gepetto wrote:Fail. Back to mod school.


Wha?! I must not be focusing hard enough...
or my mod power might not be strong enough for this one...

[Moonie211:This thread has been stickyfied and moved to the "General Video" secton!]

I bet i did it this time!
ImageImageImage
ImageImageImage
User avatar
Moonie
Give me some candy!
 
Joined: 17 Aug 2003
Location: Garland, Tx
Status: HP:11900 Strength: 940 Gold:60000 MP:1200

Postby Gepetto » Sun Oct 22, 2006 7:55 pm

That's two. One more and you're out.
And God spoke unto the Chicken, and He said: "Thou shalt crosseth the road", and the Chicken did cross the road, and there was much rejoicing.
My DeviantART profile
User avatar
Gepetto
Mr. Poopy Pants
 
Joined: 10 Jun 2004
Location: The Tokyo Settlement
Status: Bored to tears

Postby DJ_Izumi » Sun Oct 22, 2006 8:57 pm

I've actually found that to get a 'true' looking lipsync for straight up dialogue, you just wing it, throwing in mouth movements and then fixing up what doesn't look right, isntead of doing it all from scratch. This is because when they voice an anime they do NOT actually match a mouth movement to every syllable and pause, so you get an equal looking result.

I honestly think that when you do match every syllable it looks over worked and unnatural.

...But that was what I learned doing trailers. Whole nother animal when singing. Miss one syllable, sound or pause, looks like you have a robot randomly flapping its lips to the lyrics. Totally unnatural looking. x.x
Image
User avatar
DJ_Izumi
 
Joined: 03 Oct 2001
Location: Canada

Postby downwithpants » Sun Oct 22, 2006 9:17 pm

some more advice:
http://www.animemusicvideos.org/phpBB/v ... p?p=755902

my input:
synching lips is only 50% of the recipe to making the character look like she's vocalizing the words. the other 50% is body language.

also, it's generally not a good idea to show a stationary lip synch scene for more than 5-7 seconds.



oh yeah, if you're interested in becoming a mod, speak with paizuri. we could use an extra pair of eyes.
maskandlayer()|My Guide to WMM 2.x
a-m-v.org Last.fm|<a href="http://www.frappr.com/animemusicvideosdotorg">Animemusicvideos.org Frappr</a>|<a href="http://tinyurl.com/2lryta"> Editors and fans against the misattribution of AMVs</a>
User avatar
downwithpants
BIG PICTURE person
 
Joined: 03 Dec 2002
Location: storrs, ct
Status: out of service

Postby AMV_4000 » Sun Oct 22, 2006 10:06 pm

DJ_Izumi wrote:I've actually found that to get a 'true' looking lipsync for straight up dialogue, you just wing it, throwing in mouth movements and then fixing up what doesn't look right, isntead of doing it all from scratch. This is because when they voice an anime they do NOT actually match a mouth movement to every syllable and pause, so you get an equal looking result.

I honestly think that when you do match every syllable it looks over worked and unnatural.

...But that was what I learned doing trailers. Whole nother animal when singing. Miss one syllable, sound or pause, looks like you have a robot randomly flapping its lips to the lyrics. Totally unnatural looking. x.x


normally i would disagree with anything you say just to be a prick... but i have to agree with you... this is something i didnt like in videos such as Quid Pro Quo... they look like robots... like puppets... but... hey, to each their own....
User avatar
AMV_4000
 
Joined: 10 Apr 2002
Location: USA

Postby Scintilla » Sun Oct 22, 2006 10:16 pm

AMV_4000 wrote:
DJ_Izumi wrote:I've actually found that to get a 'true' looking lipsync for straight up dialogue, you just wing it, throwing in mouth movements and then fixing up what doesn't look right, isntead of doing it all from scratch. This is because when they voice an anime they do NOT actually match a mouth movement to every syllable and pause, so you get an equal looking result.

I honestly think that when you do match every syllable it looks over worked and unnatural.

...But that was what I learned doing trailers. Whole nother animal when singing. Miss one syllable, sound or pause, looks like you have a robot randomly flapping its lips to the lyrics. Totally unnatural looking. x.x

normally i would disagree with anything you say just to be a prick... but i have to agree with you... this is something i didnt like in videos such as Quid Pro Quo... they look like robots... like puppets... but... hey, to each their own....

I think it depends on how flappy it looks... which can sometimes result from having only one frame of a position before moving to the next, as was already mentioned above, or from closing the mouth after each syllable, which was also already mentioned above.
I don't know if Mckeed still pays attention to the fora, but to use him as an example, I remember commenting in 2004 how his lip synch videos always looked too flappy, because the mouths just moved too fast... however, I saw his latest Naruto parody at AnimeNEXT this year, and it still had that same look to it.
ImageImage
:pizza: :pizza: Image :pizza: :pizza:
User avatar
Scintilla
(for EXTREME)
 
Joined: 31 Mar 2003
Location: New Jersey
Status: Quo

Next

Return to Video Software Help Archive

Who is online

Users browsing this forum: No registered users and 1 guest