How are these kinds of demoic voice effects made?


Senior Member
I just started playing this game and was quite impressed by the sound effects of the voices in the intro and in the game. I'm curious how you would approach setting up an fx chain to get such effects. Any ideas?

From past experiments and googling I can say that among the things that I would test are these:
- NI Razor used as a vocoder (probably as a send effect instead of insert)
- pitch shifting with Reaper's ReaPitch (down a lot and back up a bit less seems to introduce interesting artifacts)
- NI Reflektor with some reversed IRs to get weird echoy effects
- reverb

Do you know any good tricks to add sub harmonic frequencies? Lowender is the only plugin I know of, that does something like that, but I don't have it. Does it ever go on sale? Are there any alternatives?


my office these days
The list of things one can do to the human voice is probably endless.

In this case I believe I hear some distortion, some subtle vocoding, a slight delay, some pitch shifting, and maybe a light reverb.

Really, just use what you have on hand and start mutilating a voice-over. It is fun!

If you are in a real rush check out Krotos - they have tools that do all the processing in one plugin.

Well there's 2 voices:
1 is the original pitch with distortion compression
2 is an octave below with formant shifting, could be where the vocoder comes in. (which would also make sense from a mixing standpoint, as vocoder "clean up" the harmonics)

I'm also hearing some reverse reverb near the end.

Very cool sound overall!


Senior Member
Great replies so far, thanks a lot!

Almost certainly one of the Krotos plug-ins. They seem to becoming an industry standard.
I've not had a chance to listen to the WH40K sounds yet but I think something like Dehumaniser from Krotus may be the thing to try.
I checked out the website but it's waaay too expensive for me for just a little experiment (and uses iLok?). Also I thought the examples in their sound kinda cheezy. The animal voice stuff seems cool though.

Really, just use what you have on hand and start mutilating a voice-over. It is fun!
This is what I'm doing right now and it is fun indeed!

I'll post a clip of how far I got. I think the biggest improvement would be having a better voice as a base. In the reference that I linked there is much more left of the natural voice than in my clip. But that's not really an option.





Senior Member
That sounds good Martin. What was your chain there?
Thanks! I'm glad you like it. Would a Reaper project be of any use to you (or anyone else)? I've used quite a lot of effects.

I have one track with these (ignore everything with unticked checkmarks, those are remains from my experiments):


from there I have 3 sends:
reverb send
reverse reverb send
parent track folder (both reverb tracks also feed into that)

reverb send has these:

highpass, reverb, reduce stereo width, another highpass and shelving down ~1.5khz and upwards

reverse reverb has these:

a reversed IR in guitar rig reflektor (1.9s low studio - synthetic, reversed, 77% decay, all wet and no dry signal), negative track delays to make it fade in before the actual voice (makes playback laggy as hell though, over a second latency!), compression/saturation, eq, stereo widening and pulling the sound a bit closer with panagement

And the parent track these all feed into has only one effect: ozone 7 elements, preset "low end accent", with 100% dynamics and -6db maximizer threshold. That one makes a hell of a difference, and I don't think ozone 8 elements has the same features as 7 to replicate the effect.

The core of the effect is the ReaPitch plugin. The first one in the stack doesn't do much, but the second and third look like this:



First pitched down 6 semi tones 100% wet, then pitched up 4,3 and 2 semi tones 100% wet+dry (+4 semi tones at same volume, the other two lower with +2 semi tones at -12.5 db).
Pitching down and up again introduces some interesting artifacts that I like for these kinds of effects.
Also noteworthy is that the source audio is a female TTS synth voice that was pitched down -4 semi tones on the .wav clip, so it uses the default pitch shifter for the project, which is elastique pro 3.2.3 with parameter "normal".

Hope that helps! I can upload a reaper project without the wav files if you want.