Techniques of sound synthesis

Synthesising sounds

We’ve mentioned sound synthesis in a more rudimentary form last week, when we mentioned the WDR’s interest in tools for generating electronic sound.

In reality, in the early days this looked like a collection of tone oscillators and noise generators, devices originally used to test and calibrate audio and broadcast equipment, that were instead combined in different bespoke ways to create new sounding instruments. Over time, different approaches to synthesis were codified and devices began to be produced to realise specific techniques of synthesising electronic tones.

The instruments that evolved alongside these techniques are interesting in their own right, and some have found a new life in recent times – such as the Ondes Martenot, a simple oscillator instrument that has found renewed favour with contemporary artists such as Radiohead, Gaudi, and Daft Punk. But the techniques of synthesis speak to different ideas of how electronic tones could be created, and so these are the aspects of electronic music that we will focus on.

Additive synthesis

Additive synthesis is one of the earliest methods used to try to create a complex sound spectrum. Here multiple oscillators are used to layer tones on top of one another, creating a sound with a significantly complex spectrum. Of course, this technique is limited in its usage by the number of oscillators available to be used and coordinated. A move into the digital environment has made this technique far more viable, with hundreds of oscillators able to be created instantly inside of most software environments, but even in the early days of electronic music the idea of combining different sound sources to create a new sound object was a critical element.

Subtractive synthesis

Subtractive synthesis takes a significantly rich sonic spectrum, such as white noise, and carves out parts of this with filters to create a clearly differentiated sound object. The issue here is similar to that of additive synthesis, in that you are inherently limited by the number and kinds of filters you are able to obtain and connect in your studio. Nonetheless, the use of filters on complex sound sources is clearly evident in the music we’ve listened to so far, and it creates a strong juxtaposition to additive synthesis.

Amplitude modulation, ring modulation, frequency modulation

The modulation methods of synthesis are a little different from the previous techniques we’ve looked at. When we take two synthesised waveforms, we can combine them in different ways and, using the power of math, essentially adding or multiplying them together, this will create a rich timbre depending on the frequencies of these two waveforms. Amplitude modulation combines these waveforms with respect to their amplitude, and frequency modulation combines these waveforms with respect to their frequency. Both of these methods create rich electronic timbres with far less studio equipment – instead of needing fifty of oscillators, now you only need two!

You can see some demonstrations of amplitude and frequency modulation here:

Ring modulation is slightly different in that it’s a form of amplitude modulation but differs in some key mathematical ways. This produces different harmonics and an overall thicker texture. As a result, ring modulation is often used to process recorded sound – you might have already encountered ring modulation as an effect in many places without noticing it!

The Dalek voice from Dr Who is probably the most well-known example of ring modulation.

We will talk more about Dr Who in the next lesson when we discuss the BBC Radiophonic Workshop!

Over to you

Which of these synthesis techniques have you encountered before? Which are new to you? Can you hear any examples of these in the listening examples so far?