Spiderman's "Electro" Voice? Approaches to Character Sound Design in Reason?

[Psuper]

Title says it all. If you don't feel like reading any further, please help me understand how I would approach recreating his voice effects within Reason. Here's a good example:

(from 19second to 27 seconds ' "Who are you"? and "I should kill you")

I've been away from reason for quite awhile (since reason 3 to be exact), and now that I have more time I'm going to focus on voice design as a project to re-familiarize myself with Reason in general. Particularly because live sound was impossible back then, and the software has matured greatly in that aspect.

My goal is to be able to have some core tools inside reason to work with vocal character design, stuff for movies, games, etc. Frankly I don't know where to *really* start. I've done vocoding, delays, distortion, etc and can do some basic stuff to altar voices in ways most people would immediately think about, but I can't shake the feeling I'm missing something.

Anyway, I wanted to get this thing written in hopes I can get some external brainstorming going from people who are more familiar with what is possible in the newest versions of Reason.

[VHS]

gave it about 3 minutes with a very low quality vocal sample, idk how close i really got but maybe its a start for you.


-

[marcuswitt]

@the OP: Interesting example of a mangled human's voice. To reproduce this I suggest to try two different things. First of all put one instance of Neptune in the insert slot of that Audio Channel, which is 'fed' by the Audio Track that contains the original recording of a human's voice that you want to alter. In Neptune deactivate its Pitch Adjust parameter and focus on the parameters Transpose and Formant. For instance, set Pitch Adjustment to about -8 to -12 and then Formant to +6 to +10. Such settings might produce that kind of artifacts that you're looking for. The second way of achieving it is to try the Rack Extansion Bitspeek from Sonic Charge. I'm sure that by using that geeky device and playing around with its parameters you'll get a proper result.

Maybe there are further ways of reproducing that kind of effect that was shown in the example that you've referred on, i.e. by using extreme Transpose and/or Time Stretching values applied to an audio file in Reason's sequencer. Although I haven't tried that method yet I can imagine that it also gets you where you want to be.

I hope this helps. Happy mangling!

[Stranger.]

???

[Psuper]

Appreciate the feedback guys! I do have bitspeek (got it in the May discount) but haven't messed with it much. I personally feel that with enough know-how, Reason would be the ideal platform for this type of production even though it's not specifically designed for it. I do feel I've a LOT to learn being away for so long.

However, this may be a good time to also ask if there's other programs outside of reason that focuses on or excels at sample manipulation, specifically in the area of voice/vocal manipulation. Some way you could load up 2 samples and "meld" them together, with the ability to take characteristics of each and play around with them for inspiration.

[Loque]

Vocoder. Try build in or Parsec.

[LABONERECORDINGS]

Will have a play and listen

[Raveshaper]

First of all, look into a Windows freeware called Composer's Desktop Project (CDP). It's a very powerful software for audio manipulation but it's a little bit on the older/obscure side.

As for the voice design, I think it's mostly a combination of three parts.
The top appears to be a high pass with noise added that is being sent through either a gate or LFO with a high rate.
The dry voice has something on it too but it's harder to hear.
The bottom end could be a sub oscillator being sent through a vocoder, or it could be the dry vocal pitched down or boosted in the lower frequencies and then low passed.

These are all just guesses, but I admire your choice of doing a more in depth sound design exercise as a way to familiarize yourself with the new tools.

[selig]

There is definitely a pitched down version with the main vocal IMO, probably an octave (or even one octave AND two octaves down). But also don't underestimate the contribution of the voice actor, who is also likely manipulating their voice to further the end effect.

[VHS]

Appreciate the feedback guys! I do have bitspeek (got it in the May discount) but haven't messed with it much. I personally feel that with enough know-how, Reason would be the ideal platform for this type of production even though it's not specifically designed for it. I do feel I've a LOT to learn being away for so long.

However, this may be a good time to also ask if there's other programs outside of reason that focuses on or excels at sample manipulation, specifically in the area of voice/vocal manipulation. Some way you could load up 2 samples and "meld" them together, with the ability to take characteristics of each and play around with them for inspiration.

You're looking for this: http://kyma.symbolicsound.com/
The demos only scratch the surface of what this environment can do.

[aRiver]

del

[Psuper]

I've been making great progress in the few days I've had to work in Reason. Primarily layering the sound via the merger/splitter and experimental processing each 'layer'. I've also discovered (after sampling and studying the voice samples) that there's 'breath' control applied sparingly to give the character vocal life and realism, not very obvious at first but it's there. The breath is an electric-ish noise that is subtle but powerful.

The original effects are very clean without artifacts except in the way the designer intended, and it's tricky to pull off without sounding like most of our attempts would. I'm getting close though.

So this brings me to my next question which I briefly mentioned in my prior post (and it's kinda difficult to explain what I want to do).

I'd like to take 2 samples (as an example: Voice + Thundercap), and mesh them together with traits of one imposed on the other.
So for instance, we apply the variable volume across the entire Thunderclap sample onto the entire Voice sample. Then we take the 100-200 frequency range sound of the Thunderclap and and replace the 100-200 frequency range sound of the Voice. Then we do a "chop" when we play both samples and every 5ms the sound swaps between the Thunderclap and the Voice. Basically selectively morphing the characteristics of 2 or more samples.

I know there's no single rack that will be able to do this, however I'm sure some if it is possible creating a template that can handle some of those morphing ideas at least. I'd love any help or direction on how to achieve some of this.

[LABONERECORDINGS]

Had a listen to this again, seems like there could be pitch shifting as well as AM 'gapper' style effect to give that gravelly style of vocal (listen to Star Wars as the X-Wings go through the Death Star shields, the radio comms uses AM style effect there to 'jitter' and 'stutter' the signal giving that strange metallic style raspy sound, as another reference point)

Will have a play with this, if we come up with a module / combi we'll post it up