The completed animation, drawings, just the script… what?
Probably storyboards. A rough idea as to what the scene is supposed to look like.
Usually they’re in separate sound booths with the scripts in front of them, although for the most part they have their lines memorized. The animation is then drawn to the voices, not the other way around. They aren’t really concerned with the animation because as far as they’re concerned, it’s like a radio show. They need to keep it compelling or funny just on the merit of their voices alone.
DooWahDiddy has it right.
Read Mel Blanc’s autobiography “That’s NOT All, Folks!” for more detailed info.
The entire film is made beforehand, right down to the last full stop on the credit sequence. Music is added, too, and all that’s missing is that when the characters speak, there’s no sound. The actors have to then lip-synch their interpretation of the script to the arbitrary animation version. Furthermore, they only get one shot at it. The film is played and they have to act along with the action.
This is why over 3000 actors are usually used for any one role. For example, Ellen Degeneres’ role as the absent-minded fish in “Finding Nemo” was originally performed, simultaneously, by 3557 different actresses, all watching the movie from soundproof booths, wherein they could act to their hearts’ content and be recorded 3557 different reels of tape. Most, predictably, failed to even lip-synch properly. Others managed it but got caught out by unusually timed animation or by the animators’ having banked on a Midwestern accent when they’re acting with BBC English. As the film rolled on, more and more dropped out, until at the end, as usual, there was only one actor left. In the case of Finding Nemo, astonishingly, the actors who managed to synch perfectly were EXACTLY the same ones whose names had already been written on the credit sequence, so there was no need for an expensive remake.
Not really.
On the DVD of Spirited Away, there is a “making of” special that was shown on Japanese television, which clearly showed the actors recording their lines while watching the finished animation. (This were the Japanese-language voice actors.) Do they commonly do animation the opposite way (animate first, then match voices) in Japan? In America they do record the voices first and animate to them (except in the case of dubbing something orignally made in another language). Has it always been done this way?
This is the operative phrase here. “Mrs. Doubtfire” showed Robin Williams trying to lip-sync with a finished cartoon, but that doesn’t mean that’s how it happens.
It just makes a better visual.
Not to get off topic but it’s funny how the actors act with their whole body when doing the voices, probably can’t help it…I’ve seen this in “Making Of” type shows for cartoons.
Body language influences verbal language (and vice versa). Swinging arms and pacing can be the result of certain feelings. This can set the mood for how you want to talk.
From what I have seen of several “making ofs”, voice actors act according to storyboards. Animations are then drawn to how that person acted. Body language does influence this.
THESE were… I guess that’s why I are not a college graduate.
As far as I know, it is common to have the VA try and approximate the lipsync in most lower-budget animes. You tend to see true lipsyncing in feature films and series with a large budget.
I think I saw the same Japanese TV special. There were some pretty extensive pieces of footage showing the voice actors performing in front of a large theater-sized screen showing the finished movie. It looked pretty legit.
I’ve done voice-overs, and it is damned hard–harder than regular acting, as you have no other performers to “bounce off of.” I can only speak for the projects I worked on, which were a long time ago and not high-budget: but I was stuck in a hot, stuffy booth with a mike and a script and had to read my sides over and over until the director was satisfied.
It sucked, I hated it. Oh, and Yeardley Smith is terrible at it.
Tom Hanks, in an interview, talked about doing endless takes of Woody’s voice for Toy Story, with different inflections and the like. They took the ones they wanted from the “bank” of available ones.
Disney apparently often had actors go through the motions and filmed the results, so they knew what their characters were doing, and to giv input to the animators.
I doubt if Jay Ward productions ever did this. I’ve read that they would play pranks omn each other – like setting the script on fire so that narrator Bill Conrad had to read really fast to get his lines out before they were consumed.
Almost all the voice work is done before the animation (so the animators can choose the take they like, and animate the picture to match.) However, once the film is done, the actors may come back to do ADR (automatic dialogue replacement) to change a few words here and there, add grunts and groans, re-do a take with messed up audio, and so on. For these, the actors have to watch the screen and synchronize themselves with the animation.
Although it’s better for sync to record voice actors first, that doesn’t always happen. (Shrek apparently was dubbed after it was mostly animated, as they show the actors reading to the animation during the HBO credits.) It also depends on the country. America exports a lot more media than it imports, so Americans not only expect things to be in English but to look like they’re meant to be in English. Whereas there are Japanese-dubbed versions of Simpsons, South Park, Ace Ventura Pet Detective, that Pearl Harbor movie, etc. Seeing that much dubbing, the acting becomes more important than the lip movements. Which means both that post-animation dubbing is more common in Japan, and that their actors aren’t distracted by trying to closely match the dialogue to lip movements.
I just saw Ichi the Killer last weekend. Plotless film, great dubbing. Perfect choices for the voices of the characters (despite being British accents on Japanese characters.) Once you’re into the movie, the lip sync problem goes away.
So, they could be watching finished animation, partial animation, animatics (storyboards set to time), a photo of their character or a spot on the wall.
No way Robin Williams was lip synching anything as the Genie. I bet he didin’t even have a script! The studio guys probably told him, “Do whatever”.
Generally speaking, American animation (and most of the rest of the world) will do the voices first, then animate the characters to match. Japan, however, has a longstanding tradition of doing the animation and the voices separate, probably for budget and/or time reasons. There’s usually no effort made to have the lips match the words; all that’s needed is for the animators know how long a specific speaking sequence will be, so they can make the lips “flap” for that length of time.
And IIRC, Robin Williams mentioned in interviews before the release of Aladdin that most of his dialog was improvised. I’d like to see the stuff they didn’t use when Disney finally releases the two-disc DVD edition of the movie…
From what I’ve been able to gather from Wil Wheaton’s website (he’s been doing Aqualad on Teen Titans and some other voice over stuff), he is in a soundproof room with a microphone and a script. He reads over and over until the director is happy.
[simpsons] “Animation is rarely done live. It’s a terrible strain on the animator’s wrists.” [/simpsons]
Something of a myth about Robin Williams is that he ad-libs everything. He doesn’t, he does get alot of lee-way, and he does play around, and with the genie in Alladin it made good PR for Disney to say that they just turned him loose in front of a microphone, but rest assured, he had a script. The director just chose the funniest riffs that worked within the script.
Ross, I like your style!