Found Sound & Sampling
How To Create & Process A Kick drum From Any Sound
In this tutorial we demonstrate how to create a kick drum from any sound. This would be a simple task if we started with a transient rich recording so n this instance we will start with an ambient recording from a sound map.
To make this synthesis task easier we will be using some metering plugins to help us make logical decisions:
- Smexoscope is a free oscilloscope for monitoring the waveform over the time domain.
- Voxengo Span is a free spectrum analyser with a extremely useful ‘low frequency inspection’ mode.
Here is the step-by-step thought process used to create the kick drum.
- Add the found sound to a simpler or sampler device
- Select a suitable sample start point, in the case of a kick drum we shall ignore zero crossing points as having a bit of a click at the start of the sound will be handy for producing a kick transient.
- Create a sustained frequency – this can be done using a high Q bell curve using the EQ8 device to force a strong resonating tone if there isn’t one already present in the signal.
4. Reference a kick drum to check what changes need to be made. In this case we need to transpose the pitch down to around 40-60Hz. Because the transposition on our sampler does not allows us to go low enough we can also implement a frequency shifter device to get us the rest of the way down to around 55hz.
5. We can now think about shaping the ADSR envelope to sound more like a kick drum. This is fairly self explanatory but involves removing all of the sustain to effectively give us a simple attack/decay envelope. We can then alter the decay time to trim the length of the resonating tone.
6. Next we must add a pitch envelope to the kick drum to give it the ‘thud’ and punchiness that we are looking for. By analysing a reference kick it is clear that most kicks have a very sharp drop in pitch at the start of the signal. The amount of semitones coupled with the pitch decay time will give us a range of different flavours of kick drums, so it is well worth mapping these controls to a macro control.
7. We can now Split the signal. In this case we will use three multiband dynamics processors solo’ed in parallel. This allows us to treat any problems occurring on each of the frequency bands separately. In this case we will reduce the ringing in the higher band, compress the mid band for more punch and add some gain to the sub band for more power.
8. Now we must glue the signal back together into a single cohesive sound instead of 3 split frequency bands. We will also add further processing to enhance the sound such as: saturation to add extra warmth & harmonics, overdrive to add more interesting texture and tone to the kick, equalisation to pull out any muddy frequencies and to enhance the main elements of the kick (power, punch, knock and presence) and finally a round of compression & limiting to bring the sound up to a suitable level.