I need help with a starting point. I’m a complete audio technical terminology noob but I am in the middle of a big game project and the responsibility for all of the audio effects have fallen to me. Specifically, right now I’m trying to replicate effects like you hear in these clips on clips of my own voice or others’ voices:
I don’t know if Audacity is even capable of this, but I think with all of the plugins available for it that it would be. Now this is where my audio terminology ignorance comes into play. I have clicked through all the effects in Audicty that come with the vanilla install and none of them seem to do this. So I installed the LADSPA plugins and boy wowser are there a lot of options in there.
Start with the acting. We can’t do acting. No software is going to turn your mum into Professor Festerwound.
We also can’t change the pitch of a voice. You would think it would be a snap to turn your Billy Baseball voice into a radio announcer, but nope. Can’t do that either.
Given you have a properly acted voice and it’s already in the ballpark, I would make a copy of the voice on a second track and Pitch Change the second voice down and then mix that with the main one. Then add reverb to either voice.That will give an unearthly dual-pitch effect. Control-D will duplicate any selected track.
Really Special special effects can be done with the Vocoder tool where you combine your voice with, say noise and create talking surf or talking rain. Then mix that with the other two. All this stuff is under the Effect tab.
Layer it so you can change the voices as needed, and back up each voice so you don’t get stuck not being able to back your way out of an effect because of the layers.
Never do Production in MP3. MP3 is a delivery format meant as the end of the show, not a step in the middle. Use WAV for everything. and write down what you did. Our Hollywood joke is most of producing a feature movie is bookkeeping.
If you get up close to a directional microphone, you will get the proximity effect which makes your voice deeper and ballsier, but don’t get too close or you’ll start popping your consonants.
‘Everybody Knows’ Doctor Festerwound’s voice has to be low and menacing, but it might be a hoot to get your mum to do it and then pitch up instead of down. One of the better Macbeth presentations had the three witches instead of flouncing around the stage in black rags, just three ordinary-looking women that only when they got together turned into something truly evil.
The Audacity vocoder hasn’t to hide itself (except for the execution speed perhaps).
Simply use it with the highest amount of bands (240) or enter even a higher number (1000 for example).
Trebors example can be made by using the built-in “Radar Needle” which is nothing but a buzz tone (x multiples of the frequency with equal amplitude).
Set the radar needle amplitude very high (the other ones low or zero)and choose a frequency of about 80 Hz.
Other sounds that are best suited for the right channel:
Square or saw tooth (low frequency for more overtones, varying or constant)
Fire and explosions
Hammond or similar organ pads.
Phase/Frequency modulated sounds
Applause, cheering audience
the original voice doubled with delay, reverb, pitch shift or paul stretch…
…and so on.
Increase the white noise amplitude if the speech is not understandable enough and/or keep the original voice track to mix it in.
If the music signal is too simple, you can convolve it with noise to broaden the signal:
Type in the nyquist prompt (for mono tracks only)
(convolve s (mult 0.01 (abs-env (noise 0.1))))
By the way, I prefer the distance set to one in order to keep most of the original voice without disturbance.