top of page
Blog: Blog2
  • Writer's pictureAsh Saron

Synthesis


How synthesizers work

Now we understand the theory of how sound works, and how different instruments produce it in different ways, we know enough to build ourselves a synthesizer. You can probably see already that a machine that can copy the sounds of virtually any other instrument would need to be able to:

Generate sound waves of different shapes.Generate more than one sound tone at once to produce a fundamental frequency and harmonics.Make the volume of the sound change over time to produce different ADSR envelope shapes.

That's pretty much what an electronic synthesizer does in a nutshell. It has a number of different voices or oscillators (sound tone generators), each of which can produce waves of different shapes (sine wave, square wave, saw tooth, triangular wave, and so on). It can combine the waves to make complex sounds, and it can vary the way the sounds attack, decay, sustain, and release to make the sounds mimic existing instruments like pianos.

To make a synthesizer sound somewhere between a piano and an organ, you could select a square wave generator (which gives an organ-like sound) and set the ADSR values to be like those of a traditional piano (slowish attack, quickish decay, long sustain and release). Modern synthesizers have "presets" (ready-programmed settings) or "modes" that let you select particular instruments at the flick of a single switch. Of course, you don't have to copy traditional instruments with a synthesizer: you can change the settings to whatever you like—and create all kinds of sounds no-one has ever heard before.


Adding or subtracting?

In art, there are two ways to make a piece of sculpture. You can take materials you've found in the world around you and stick them together to make something completely new. That would be an example of working in an additive way. Or you can start with something like a big block of stone or wood and chisel it down, slowly reducing it to what you want by stripping bits away. That's working in the opposite—subtractive—way.

The same is true of making sounds with synthesizers. It's perfectly possible to build up a complex sound from simple tones that you add together and shape in various ways, which is more or less the approach I've described above. But it's much more common for real synthesizers to work the other way, through subtractive synthesis. That means you start with a complex sound, filter it to remove harmonics, and envelope shape its volume. In practice, then, a simple subtractive synthesizer makes sound using four independent components:

An oscillator generates the original sound (and you can control it in various ways), which will be a mixture of a fundamental frequency and its harmonics. Most synthesizers have at least a couple of oscillators.A filter cuts out some of the harmonics (for example, by boosting or cutting all harmonics above a certain frequency).An amplifier changes the volume of the sound over time, according to ADSR values that you set (as we discussed above).A second, independent oscillator, known as an LFO (low-frequency oscillator), can be used to vary how the previous three stages work, producing some very interesting effects. For example, if you apply the LFO to the original oscillator, it makes the sound it generates wobble about in what's called a tremelo effect.

Yes, this is all sounds a bit abstract and mathematical. It's easiest to understand what it means in practice by experimenting for yourself; at the bottom of this article you'll find some suggestions for apps you can download that let you play around with your own simple subtractive synthesizer.


Analog and digital synthesizers

The original synthesizers achieved all this using laboratory-style electronic equipment that generated and manipulated actual sound waves. Instruments like this are known as analog synthesizers because they work directly with the sound waves themselves. Many of these synthesizers had lots of separate, sound-creating modules that could be connected together ("patched") in different ways; that's why they were called modular synthesizers.


From US Patent 2,855,816: Music synthesizer by Harry F. Olson and Herbert Belar, courtesy of US Patent and Trademark Office.

The RCA Synthesizer: Back in the 1950s, the world's first analog electronic synthesizer was programmed using reels of paper tape, punched with holes that represented things like the frequency, octave, envelope, and volume of the sounds required.

Modern synthesizers do everything digitally, by manipulating numbers with computer chips. Not surprisingly, they're called digital synthesizers. They're essentially computers that have been specially programmed to generate and manipulate sounds. Most synthesizers can be connected up to personal computers, so the computer can be used to store and record the sounds the synthesizer makes or play it automatically. To make this sort of thing easier, computers and synthesizers use a standard way of connecting together known as MIDI (Musical Instrument Digital Interface).

Another kind of digital synthesizer, the sampler, lets you feed in a recorded sound (maybe the noise of a sparrow singing) and then manipulate it in various ways by changing the sound settings. So you can make the sparrow sing more quickly by speeding up the sound, or play the bird-song on your keyboard, so the low notes sound like older, heavier birds and the high notes like younger, smaller, and chirpier ones!

bottom of page