In this article for Production Expert, Tom Lowe discusses his use of Krotos' Dehumaniser 2 software for helping him create monster sounds for a pharmaceutical company’s advertising campaign.

I first became aware of Dehumaniser 2 a few years ago and have since read several articles about its use on Production Expert. When I first became aware of it, I didn't have much use for the software but was aware of its capabilities and impressive results. Fast forward to a couple of months ago when I was approached to create the sound for a video featuring a monster that was advertising a new product for a pharmaceutical company. The product is for asthma patients, and the ad agency had developed the concept of this monster to visually represent the feeling asthma patients go through.

My thoughts immediately turned to Dehumaniser 2 as a tool that could help design the sound of this character. So it was bought in quite early in the stage when a storyboard and animatic had been completed. The idea was that I would start developing the sound, and the character design would evolve simultaneously, taking the way he sounded into consideration. Usually, on projects like this, I'll be bought it once all the animation has been completed or in the final stages of rendering, so being brought in relatively early in the process made a nice change.

Capturing The Performance

One thing that must be right when creating character voices is the initial vocalisations. You could have every sound design tool available, but if the initial performance for the character isn't right, it's going to be much harder to get a result you're happy with. As character voices aren't one of my strong points, I asked a colleague to perform these for me and record them in his studio. I've seen that a popular way to work with Dehumaniser is to record directly into a track with the plug-in instantiated to effectively perform directly into the plug-in. Suppose I didn't have the luxury of knowing the animators would be using these vocalisations for lip sync and to alter his gestures accordingly. In that case, I'd have certainly considered working in that way. However, we were fortunate enough to have rough playbacks of the early stages of animation, which were used when recording. This meant I didn't have to edit the timings of the vocalisations to fit the animation and could concentrate on altering the voice.

Once the voices were recorded, I started playing around with various settings to get something ready to send to the client. There are several presets to choose from, which can be a great starting place to get a sound you're happy with. However, I decided to start from scratch. Although this character is a monster, there is a definite cartoon element to him, and I wanted to keep any "words" he mutters distinguishable and not altered beyond recognition. The other thing I wanted to try and emphasise is the heavy breathing/wheezing in his voice. Luckily, the performance had achieved some of this, giving me a great place to start from. Using the Scrubbing Convolution module to trigger an additional sound layered on top of the original sound. The result was subtle, but something I felt undoubtedly added to the overall effect. The fact that it reacted to the original signal level rather than being applied uniformly was something I felt made it a very useable tool. It kept everything sounding like it was coming from the same character, adding to the voice rather than completely changing it.

Flexible Routing

After using some modules to get a sound I felt was heading in the right direction, I started tweaking some settings. The ability to change the modules' order, split the original signal, and have parallel modules used for processing was something I found incredibly useful and effortless. Despite the power of Dehumaniser 2, the UI is not overwhelming, and it's effortless to get a sense of what each module is doing to the sound. Changing the order of modules to get the desired result, such as pitch shifting the signal after the Scrubbing Convolution module instead of before it, was an excellent example. The change in overall sound was just right as a result. Also, the ability to keep the same settings but alter the sound by changing the order of modules was something I found helpful and a great way to tweak the sound without undoing any previous work.

Dehumaniser 2 can create a wide range of effects, from those that completely change the sound beyond all recognition to making more subtle changes, which shows how versatile it is as a sound design tool, which is very appealing. The result of all this was a happy client and me discovering a new plug-in that has now become part of my sound design arsenal.