Wadjet Eye Games

Author Topic: Robotic SFX / sound design  (Read 3544 times)

Offline lawroboros

  • Newbie
  • *
  • Posts: 1
    • View Profile
Robotic SFX / sound design
« on: August 16, 2013, 09:55:06 AM »
Hi,

I'm absolutely loving the game (not quite finished but not long to go by the looks of things).

As a games industry audio techie I'm incredibly impressed with the effort and detail put into the robotic vocoder effects used on the dialogue throughout the game - particularly Gimbal and Memorious.

I don't think I've heard anything as good since WALL-E came out; I was wondering if there are any specific plugins or techniques that were used, and whether any special considerations had to be made regarding direction of the voice artists?  For instance, it sounds like whoever voiced Gimbal spoke in a particularly querying manner.

Would love to hear more about how the above was achieved.

Thanks!

Offline bubblepipemedia

  • Newbie
  • *
  • Posts: 3
    • View Profile
Re: Robotic SFX / sound design
« Reply #1 on: August 16, 2013, 09:13:42 PM »
Thanks for noticing!

A lot of work went into the processing of voices in the game and the team was very involved in selecting exactly what quality each voice should have. I usually sent multiple versions from them to choose from. Sometimes I was sent back to work on it again. I think it was primer (and maybe another one), we ended up doing a lot of trials with before we came to one we liked.

The voices were recorded by Dave and he did the of direction, with lots of input and additional direction from Mark, Vic, and Jim. I focused on the effects the voices should have and how to manipulate them.

For the initial robot effect, I primarily used Kore 2 (plus the 'Deep' expansions) with a lot of tweaking. Later on, for some robots, I ended up using the Pitch Transformer in Logic 9 for some of the robots. The Logic plug had a good generic robot sound, but it also had such a distinct character that I could only use it on a few robots (and usually not the important ones).

I then used parallel routing to output a copy of that voice to speaker/cabinet emulators (mostly guitar rig 3) and usually another parallel rout (pre-robot voice) with a really high pass filter, so that sibilance could be heard (often the robot voice made it hard to tell certain letters apart, this helped with intelligibility, which is very important in games, especially adventure games).

On top of that, lots of expanders so that the robot voice had dynamics and followed the expressions of the original phrases. And, of course, compressors, to make it all uniform in volume. I also used a match EQ to help make sure the voices all sort of jelled together, though I only used it slightly (maybe 33-50% at most).

Shoot me an email sometime at NHC at bubblepipemedia dot com if you ever wanna talk shop!