If the animation is done first, do the actors have to remain strictly to script leaving no room for improv and also have to nail their timing?
And if the acting is done first, how do they time it? Do they have to do it blindly, only by imagination?
In: Other
Typically voice work is done at the animatic stage. Basically, the storyboard is crudely animated to give the production a sense of timing and what everything will look like.
[Robot Hell Animatic](https://youtu.be/d7h6rdrCP-I)
[Robot Hell Finished](https://youtu.be/DR7rurcamNo)
As far as sticking to the script, it really depends on the director. If the actor comes up with something better, the animatic can be easily updated before major work is done, but is is about the director’s “vision.”
Animator here for 2D at least, Lets get basics down here first:
“Preproduction” or “Preprod” is where concept art/Script/Storyboarding is done
“Production” as the name suggests is well the actual production, besides animating, anything related in making the story come to life.
Now obviously all studios have diff methods but they all follow similar workflows.
Script writing/Storyboarding are usually semi-finalized in Preprod ASMAP (changes can be made up until its released)
**Now to answer your question**
Once a script/storyboard is semi finalized, studios usually develop something we call an **Animatic**
An animatic is basically the rough outline of the whole film. Its basically the storyboard stitched together as a film with timing just based on the intuition of the people making it(if you imagined the dialogue; you can easily re enact it)
The animatic is essentially a blueprint of the whole film
**This is usually the first iteration of the Animatic**, which is what VA s work with. Once the VAing is done, the animatic is once again reworked for polishing (especially if there are improvs)
then the animatic is reworked till its ready for the “production” for keyframing/inbetweening/etc.
**Another common practice** is that they have the VAs act it out first (just with the script) then the animatic is based of those recordings(under the supervision of the core preprod team) and again reworked continuously until the studio deems it to be fit for production
3D process is similar, and more flexibile since the animatic can be done in software with stiff animations (see clone wars lost episodes)
Voices are recorded first for a lot of animation (and then the animators match the facial animation to the voice).
But I just recently watched a YouTube podcast (called “Trash Taste”), where they interviewed a Japanese anime animator (Ken Arto) who has animated on a lot of the most popular recent modern animes.. and he said that for Japanese animation, they do the animation first, because it’s ‘faster’ that way.
This is somewhat surprising, because among people who watch anime, there’s often a split between fans who prefer “dubs” (where English voice actors dub the Japanese animation) vs “subs” (where the show is presented with English subtitles and the original Japanese audio). But it turns out that, apparently, all anime is “dubbed”.
So maybe the truth is somewhere in the middle. Different animation studios do things differently.
Animation dialogue recording engineer here. I’ve worked across TV and film for the last 15 years and although the exact process differs from production to production, the preference of the production companies is usually to record the voices prior to animating. Sometimes though, due to animation studio workflow, broadcaster or studio deadlines, or other factors (Covid for instance), it’s not possible and animation begins before voice recording.
There are pros and cons to working with or without picture. Without, the actors are totally free with the timing and interpretations of their reads – within the constraints of the stage directions and the context of the line – and can be recorded either individually or as an ensemble. With picture, the actors can see how their lines fit the on-screen action – which is especially useful for adding efforts or ‘walla’ – and are recorded individually.
With a wild recording (not recorded to picture) there will likely be pickups at a later stage which might be recorded to picture. This can be at the animatic stage (which means recording to the timing of a sketched shot, but not matching lip-sync – still allowing some room for improv or reinterpretation of lines), or it can be to animation (full or previz) which means matching the lip-sync – called ADR (Automatic Dialogue Recording / Additional Dialogue Recording). If portions of a line are off screen, then some leeway is possible with timing.
With a wild recording, the audio will go through any number of edits before it’s animated. Often the first edit will be a ‘radioplay’ edit where all the lines from all the characters are sequenced and roughly timed to flow well. Then an editor at the animation production company will work with the animation director and storyboard artists to retime the audio to match the animated storyboard (the animatic).
I highly recommend watching videos of Justin Roiland doing voice takes for Rick and Morty. The best is when they get him real drunk so he can sound even drunker than normal Rick.
Check out the South Park documentary on HBO, it’s funny and show you how they make a single show in just 6 days. They show the process of voice acting and syncing with the animation.
For those who want to see how the sausage is made:
in Japan:
[https://youtu.be/MeQlB7XR2_A?t=265](https://youtu.be/MeQlB7XR2_A?t=245)
In the US:
[https://youtu.be/D1AnSUziC70](https://youtu.be/D1AnSUziC70)
The process, generally, goes something like this:
**1.** Script is written, which includes all dialogue.
**2.** Storyboards are drawn. They’re basically like big comic books that help animators know what to actually animate later on.
**3.** Animatics are crafted from the Storyboards. It’s like turning the comic books into a fancy Powerpoint presentation that they can click through at any time.
**4.** Voice actors read aloud/recite from memory the script to the “rhythm” of the animatic. They make the funny noises and voices for the drawings as they move from drawing to drawing, like a dad reading a bedtime story to his kids.
**5.** Animators take the combined resources of the animatic and the voice recordings to animate everything in sync. Usually, they’ll animate the actual mouths last, focusing more on the movement and positioning of the characters in the scene.
**6.** Finalization. Everything else that goes into making everything nice and shiny, like line work, coloring, lighting, etc.
During the voice recording phase, the VAs occasionally have the opportunity to throw in a joke or two or add lines of improv, and the storyboard artists will manually add to or revise the storyboard/animatic if the director likes it enough. Since there hasn’t been any actual animation done for the project at this point, there’s plenty of room to experiment. Sometimes (apparently more often than I originally thought), the VAs record their lines first, and then the storyboards/animatics are drawn after, basing their animation on the energy and tone of the actors themselves.
Maybe this is too simple, since I’m relatively new here, but these are the basics, and the process is generally different from studio to studio.
I was thinking about this recently—how much of Robin William’s Genie performance in Aladdin was ad-libbed? Did they just put “Do a Jack Nicholson impression when you say ______”? Or did they just give him a general script, record that as well as all of his improv tangents, and then animate the finished audio cut? I assume the latter.
You do the voices first, at least in the west. It’s acted out, then the animator times their animation to the voices and creates the character performance. In any animation, you have the sound files as a layer and scroll back and forth to find the timing of the dialogue and which drawings to put where. It used to be that you’d have to mark off the dialogue in an x-sheet, basically a sheet noting which syllables are said at which frame.
In Japanese TV animation it used to be common(probably still is) to animate the performance first and have voice actors come in after- hence the anime random mouth style of lip sync. Makes it easy to stretch if needed and does away with the timing step.
Latest Answers