4

I have recently been listening to a lot of Melbourne Bounce and I was wondering how the producers of that music make the human-voice-like leads.

So I fired up my DAW (Ableton) and started synthesizing. My weapon of choice was Serum by Xfer Records because of the fact that you can import your own audio and play it back in cycles. This meant that I had to get a single cycle waveform of the sound I wanted.

WRONG: It did not sound at all how I wanted it to. Then I thought it had to be modulation that would make it sound the way I wanted. A spectrogram confirmed my thoughts. So I moved on, trying a whole new approach.

I have some glitch vocal chops which sound like the bass from a lead like that. So I dropped one on a sampler and tried different effects. The closest I got was when I used the Amp and Saturator effects that come with Ableton, plus your everyday EQ/Compression. But the problem I got was when I played for example, E3, it would be a way shorter sound than E2, due to samplers time-stretching (or rather, it not time-stretching).

So now I am back to zero and I have decided to write on Sound Design.

How are these lead sounds made? And in the future, when stumbling on another sound I'd like to remake, how do I effectively analyze the sound so I can recreate it?

Let me know if I should break this up into 2 questions or if there is any good answers on this site already (I tried searching on this site but could not find anything).

Here is a track with the sound i'm looking for:

  • 1
    You're more likely to get answers if you include an example of this "Aah" sound... – Marc W Jan 25 '16 at 3:15
  • @n00dles I edited my question – BRHSM Jan 26 '16 at 10:17
1

EDIT:

Find a vocal sample that has a defined note and is low enough to be pitched upwards.

Take a 250ms chop from this vocal, preferably one in which there is a slight amount of pitch movement (vibrato). I recommend using a spectral view of the sample rather than a waveform view to make an informed decision about what 250ms slice to use.

Pitch the sample up to taste (probably around an octave).

Loop the sample to continue playing as long as you hold a note.

Add reverb/distortion/bandpass filter to taste.

Use a lot of pitch bends and move up and down entire octaves in you melody.


I'll leave the recipe for this other dutch house sound here...

2 oscilators

one sine one detuned saw

band pass filter with some resonance

sharp compression

reverb

pitch bend the notes

you can add some kind of stereo widening too (maybe delay with a short time setting), as well as some tremolo/vibrato for extra movement

  • One detuned saw as in one voice or multiple voices but one oscilator? – BRHSM Feb 16 '16 at 10:41
  • 1 voice = 1 oscillator, when you add extra voices to an oscillator, under the hood its just copying the oscillator and applying small changes to various parameters (usually according to your unison parameters). Practically speaking, an oscillator with 3 voices is really 3 oscillators. Anyway, one saw oscillator and one sine oscillator, though the amount of voices isn't really essential to this sound. Its more about the detuning and filtering. The most essential part of this sound is the midi programming though (bend notes). Hope this is helpful. – Alfred Young Feb 16 '16 at 22:44
  • I can't get the filtring right. Can you give some details on that? – BRHSM Feb 17 '16 at 13:25
  • I sat down to actually recreate this sound for you and it turned out I gave you instructions for a different dutch house sound. Sincere apologies. Heres the video that helped break it down for me: youtube.com/watch?v=TXQaC0_sDoA. The video doesnt give instructions on how to make the original sound but my edited answer should help. – Alfred Young Feb 18 '16 at 20:16
  • 1
    theres a free sampler called 'grace' that looks promising – Alfred Young Mar 25 '16 at 6:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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