Hello Everybody,

Ive been wondering for many days to create monster voices with emotion. So the idea is to get emotion in voices like in the movie WALL-E, but they are robotic.

So are there any ideas to create emotions such as Sadness/happiness/surprises etc... for monsters such as Dragons, dinosaurs or any fantasy creature/monster (not anger & attack growl etc...as its kinda easy & its available in many SFX libraries also).

Kindly share your ideas regarding this, it would much appreciated.

Best Regards,

5 Answers 5


Personally, I find that creature vocals are one of the most challenging things to creature. Alexandre is on the same path I would like to touch on, however maybe elaborate a little further.

To me, creature vocals boil down to two specific, but broad concepts - that of dialogue editorial, and that of vehicle sound effects editorials:

  • When it comes down to the essence of a creature vocal I find that I have to dissociate it from being a sound effect event And instead look at it as a dialogue event - instead of being English, it's just gibberish. The thing about dialogue editorial experience is that it develops one's hearing sensitivity toward the nuances of the voice which most of us take for granted. Nuances like cadence, inflection, rhythm and so forth. This all converge on the term 'emotion', because each of these characteristics, among others as well, serve as a vocal body language to express to our peers what we're feeling (such as downward pitch for sadness often, abruptness for surprise, wheezing for pain or strain, and so on). The challenge is figuring out how to lift the emotive abstract from spoken language and imprint that onto a combination of sound effects to create a voice. The translation is not an easy one, but this is where I believe having background experience in dialogue is key - because it has taught you how to properly critique the voice in abstract (as in, critiquing and analyzing the 'sound' of the voice, not the 'content' of what's being spoken).

  • Now what could creature vocals have to do with vehicles? Vehicles are a precision process. it's all about using the right combination of source, but careful to not create a "doubling up' of multiple engine types (so as to ensure the engine sounds as if it's a singular unit), and blending in and out of multiple source files at the correct pitch cross point so the move is fluid. This is what I find applies to successful creature vocals. The difference between 'blurry' creature vocal elements versus those which have had great care taken to their alignment a la a car engine edit is a night and day difference. The difference is that when all of these vox elements are careful working together as one unit, the vox feels to us as being a believable organic entity which is emotively expressive. It doesn't mean all elements of a growl should be aligned perfectly, but understanding the cadence of a growl - the pitch arc and roll from throat at the top to guttural at the tail (just one example). So much like cutting vehicle engine rev out and away, it's about selecting the right combination of elements which blend to create this arc, but have the illusion of being one sound

When it eventually comes to cutting these emotive creature situations, this is where the dialogue background comes in - by knowing what the abstract emotive quality is that you want to express, knowing through dialogue editorial experience how this emotive quality is imparted through dialogue/vocals, and thus being able to analyze source material to find what intuitively feels right. Sometimes you can find some stuff which hits the emotion on the nose, other times you may spend a few hours throwing things at the wall to see what sticks. Sometimes even, you have all the right elements, but all it takes is some careful nudging to hit that suite spot where they marry perfectly. For me, I work intuitively on this stuff in terms of the actual building - trying to reach that point where I'm connecting with what I'm hearing like sharing in the pain I feel for this creature whose dying and choking/growling lightly in the process. Sort of like cutting a fight - you try to build up your elements so that what you're hearing in the edit feels like it's hitting you square in the chest. Same sort of idea for creature vocals - when you find yourself beginning to empathize with what your hearing, you're beginning to give that abstract emotive a grounded, sonic identity.

And this building part is where I feel dialogue further comes in - being able to listen to what you've cut, and think in terms the rhythm of breathing, whether this particular moment is missing a vowel like roundish element, whether the growls sound great but stale (because maybe we need some air-like larynx texture just to help glue the percussive/vocal elements together to give it a natural air-moving vocal chord quality). Basically, being able to do a mental spectral break down your final sound in your minds eye into the parts you think you need to achieve it (air, throat, chest, etc).

So I believe dialogue editorial experience will help you to analyze vocal emotion in the abstract, and vehicle experience will help you to be able to articulate these emotive abstracts into a final creature voice.

Just my 2 cents.


I agree with Alexandre that a lot of the reason WALL-E was such a successful voice design was because the sounds were supercharged with an innate humanity.

As an alternative to finding human elements in a library, try voicing it yourself! Or perhaps find a friend with voice training (or who likes to talk to his dog a lot... you know, in dog).

Set up a mic, watch the footage (if it's a film you're designing for), pick a monster and emotion and just make a lot of sounds! Drink room-temperature apple juice to help keep the voice running. If nothing else, it's a ton of fun — one of those I-can't-believe-this-is-my-career kinda moments — and it can be very inspiring. Let your roommates know ahead of time, of course, lest they hear you through the walls while you're channelling your inner-Andy Serkis.

Even if you don't wind up using the human track, it can help you lay out a groundplan. Then you can look through sound libraries for elements and layers to build your ideal sound on top.

  • Cool idea Matt, Thanks so very much & I'll try these methods... Cheers!!!
    – Bala
    Commented Jul 18, 2012 at 15:37

Good question! I think we interpret monster emotions very similarly to human ones. Meaning when you're sad you have down ward tone inflection... excited an upwards one. We even associate these emotions to when animals make these sounds. Like a sad dog crying, or a cat meowing cause it's hungry.

My approach would be to use this as a reference when recording. If you can add a processed human voice somewhere in your layers the emotion may come out. Even if you process the voice beyond recognition, the idea is to maintain the dynamics and pitch. That is really the key to emotions.

Here are some interesting links:




  • Thanks very much alex - those are some useful links, I appreciate this. Cheers!!!
    – Bala
    Commented Jul 18, 2012 at 15:35

This classic Jim Stout tutorial should be a great help; using your own voice as the carrier signal for a vocoder seems like the way to go with this. Use this technique to manipulate a mix of animal noises and, as the others have suggested, just let loose and have fun!

  • Using plenty of emotion when you "let loose", of course ;-P
    – Skarik
    Commented Jul 18, 2012 at 20:44
  • Thats sweet Skarik - Thanks for the link - I'll attempt this one. Cheers!!!
    – Bala
    Commented Jul 19, 2012 at 12:53
  • Hello, I would love to use Stout's method in the animation that I am working on at the moment. Would you be able to explain to me how to set it up in Logic 8? Many thanks, Zuzia
    – Zuzia
    Commented Mar 19, 2013 at 8:28
  • I have answered in your other SSD question: socialsounddesign.com/questions/18606/monster-sound-design
    – Bit Depth
    Commented Mar 19, 2013 at 9:49

I think when it comed to Monster Voice sounddesign, layering is very important. At least one track for the speech that should only be changed to an extend that someone Else with a fresh pair of ears can understand it. Next to that Layer i would use Crazy animal and human shouts Growls whistkes. Especially by combining human speech with Animal Sounds should work well. If you process them the Same Way, like Stifting Formats, Same Vocoder etc you can combine them. When you use any Kind of effect from pitch, formsntshifting to destortion and grainular processing, Automate the Parameters by a huge amount to Support the phrases and emotions. For wall e they used crazy pen-Controllers to live track the formants and vocoders on those Roboters in a x y z Koordinate system. Also try to Vocode Long vowels with Animal Sounds as modulator. Gl

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.