Exercise in Sound Design for Cartoon Games
Posted by Bryan on10/12/2012
Having an opportunity to work on a cartoon project can be one of the biggest thrills for a sound designer - it gives us free range to be as goofy and frivolous as we want, and deep down we all secretly wish we were Disney "Imagineers", devising impossibly clever practical ways to make it sound like the cartoon pig's tail goes out straight when he gets surprised, while wearing smart ties with a short sleeve dress shirt tucked into our modest 1950's slacks.
Last month, I was asked to collaborate with Coatsink Software on their upcoming iOS title "Fatty", and create a voice and in-game sound events for the titular character, a cute green cartoon blob. For a sound designer with no vocal training or acting background, stepping in front of the microphone in a limited budget scenario can be daunting, but the brilliant character design inspired the goofy little kid in me and gave enough direction to inspire confidence in my own performance.
The main thrust of the character voice work was several small snippets of pigging out that came at the end of a level when the score is tallied and the player is presented with a grade. The grade is based on how much food you successfuly ate, with a visual of Fatty, either skinny, bloated, or morbidly obese and drooling all over himself, so the voice performance was split into three different "sets" to clearly delineate the different levels of achievement and character size. As the game had 10 separate ranks, the intensity of the sound montage had to be altered in 10 different ways and still have a clear progression - the risk here was relying too heavily on my own performance, starting with Level 1, and finding halfway to #10 that I had exhausted my range. To keep things evenly paced, I started experimenting with character voices at the two extremes - trying to develop a voice for the Level 1 Fatty that was as small, cute and meek as possible, and jumping all the way to Level 10, trying to get my voice as deep as possible, while grunting, drooling, and pushing the limits of human decency. Wherever a middle ground was between the two, I would have cleanly set out three groups of vocal performances, the Small, Medium and Large, and could vary these voices with additional layers of mouth noises, lip smacking, and generally more gusto, without worrying about my personal range on the way to Level 10.
Each performance only received a small amount of processing - I have a lifetime of making stupid sounds with my mouth that have come in handy in this regard - with some pitch shifting to raise or lower the formant at either extreme. The smallest voices were lightly pitched up, the middle was untreated, and the Large tracks were double and the 2nd version was lowered by about an octave, dropped the gain and buried in the mix, which offset the voice with a slight monstrous feel, and made it feel as if it were coming from a bigger sound source with deeper resonance. Each recording session for the three voices was split into long uninterrupted takes with sets of chewing, grunts, mouth smacks, breathing, and slurping sounds that would be layered on top of each other to create an edited performance. This would help make up for my lack of comedic timing, and inability to deliver long, consistent performances. Several aspects of the different performances were layered together to create the ideal sound, and since this was dealing with cartoon characters, over-exagerating the details worked in our favor. It's a bit shameless, but as an amateur voice actor you can scrape together whatever you can muster into a good performance with a little ingenuity and smart editing. Just be willing to swallow your pride in the recording booth and deal with the embarassment of sifting through the raw recordings.
While an adequate amount of humor and characterization ground work was laid down with the initial recordings and the differing levels of eating intensity were established, a balance of action could be incorporated into the montages with some nice gross food foley. Mainly functioning as a basic level of detail to tie the montage together, this was also another opportunity to inject humor and push the cartoonish style of the game, while further differentiating the levels. So a grocery list of demolishable foods was assembled with levels of textural differences. Level 1 would have an almost health conscious ingredient list - nice leafy green kale, and tasty diced apple that provided a light mix of leathery chewing sounds, and crunch without over-the-top smooshing, while the other extreme, Level 10, was layers of Jell-o, pudding, and fully smooshed bananas. In between the two extremes, the balance between the dry textures and gross glooping whould shift, and the performance would become more aggressive and violent. The big payoff was the clear differentiation and sense of transition between crushing apple, versus crushing banana. The crunchy but wet texture of apples lent itself to the more organized feel of the Level 1 stuff, and cross-fading the two together created some nice textured mouth noises. Small details like slurping Jell-o, and full mouth chewing were added with a small round of additional vocals, and then I washed my hands and wiped apple sauce off the floor.
With everything laid out so clearly in my head from the beginning, now that I had over 10 minutes of raw recordings and details to pick through and sort into 10 different sounds of 10 different levels of intensity, all I had to worry about was comedic timing. My horrible mouth noises had left me scorned by my loved ones and shunned from polite conversation, but I was officially Fatty.