This week I’d like to give a quick tip which involves a unique technique for differentiating every note of your chosen scale.  Please note that I’m going to weave in some light music history discussion to reinforce the point, but stick with me!  The basic idea is to use a different sound for each note instead of just one predictable synth tone that is being transposed.  Let’s dig in.

To begin with, you’ll need some kind of sampling device or software such as Sampler (in Ableton Live), Kontakt, EXS24, etc.  Just as long as it can easily create and play back multisamples, it’s usable.

The first bit of work is to figure out what scale you are writing in, specifically which notes need to be taken into consideration. Once the scale is decided upon, the next step is deciding which samples to use for each note. In the key of C, for example, you’ll need to decide on samples for C, D, E, F, G, A, and B.  And this is where the true creative work begins.  Hip-hop producers have been doing this to varying extents for many years since the genre was traditionally a sample-based affair.  Some of these producers went as far as to use single samples from many different pieces of vinyl, piecing together entire musical phrases from seemingly disparate bits of audio.  Of course, each sample might represent a unique note in a scale so that one could potentially put together traditional melodies and harmonies using this material.  This isn’t a new idea by any means, but it seems under-explored outside of indie Hip-hop and some of the more abstract genres.

When you go about choosing the samples for each note, a good rule of thumb is to make the sure sounds are related to one another. This relationship can be explicit, such as all the samples being brass instruments or being from a certain traditional musical form. However, you can also look for relationships that are slightly more abstract, such as all the instruments coming from found sounds in a forest. At the end of the day, there’s no set rule other than the obvious: The samples should sound good being played as a scale. These are merely suggestions to get you started finding sounds that will work together.

OK, back to the technical stuff.  Once you’ve spent some time finding audio samples that work well in combination, it’s time to load them up into your multisampling instrument of choice.  In Sampler, it’s a simple matter of opening up the zone editor, dragging and dropping the various samples on to their respective keys.  This is done by limiting the keyzone of each sample to just one note as well as setting the proper root for each note:

samples mapped to the first 3 notes in the key of C, two octaves below middle C

Some of the audio might need to be massaged into shape, for instance with high pass filtering to get rid of extraneous low frequency content or short delays to brighten the timbre and extend the lengths of the notes.  As you test out your sample-based scale, you may find that some of your chosen sounds just don’t retain their interest when repeated many times in a run of notes.  Again, choosing samples is the most creative (and potentially frustrating) part of the process, and these decisions have a large effect on the quality of the scale as a whole.  One dull “note” might throw off the entire vibe of a song, rendering it unlistenable.  I never said it was going to be easy, but when you get it right the results are very satisfying.

One great way to extend the concept of using short bits of audio this way is to use various stretching techniques so that the samples can be infinitely sustained.  This can easily be achieved using the looping functions found in most sampling software.  The idea is to choose looping points within the short samples that sound good while sustained, and this is an art as much as it is a science.  It certainly helps to have some kind of “smoothing” function such as crossfade-looping in order to get rid of volume spikes at loop boundaries.  For instance, I might take a 3 second trumpet sample and set up a 1 second loop in the middle of it.  This kind of thing has been done for years, but in the context of designing a sound for each note in a scale, it’s still surprisingly underused in non-gimmicky ways.  I’m talking about more than just using samples for ironic effect here.  Ideally, we’re crafting new timbres that might bear only a passing resemblance to the original sounds.

Even if you are familiar with most, if not all of the techniques mentioned in this article, I hope it’s at least given a fresh viewpoint on an old trick!

Thanks for reading, and until next time please stop by nickstutorials.com for video tutorials on sound design and music production in Ableton Live.