Synthesis means creating your own sounds
Timbre project:
This is an absolute
beginner's guide to
sound synthesis
top Have a little think about how to create sound

To create realistic musical sounds two sorts of generators  are needed

 1  oscillators to produce the basic waveform and
 2  envelope generators to change the waveforms as the notes age.

The oscillating medium in many woodwind instruments is a reed;  for brass instruments and the didgeridoo the oscillating medium is the player's buzzing lips.  These create standing waves.

What types of waves are needed to synthesize brass sounds?  What else is needed apart from those  waves?  Listen to these real brass instruments and describe their timbre. 


1    Basic waveforms

Sine wave: simply the fundamental with no harmonics.  Pure. Used on its own for creating the sound of a whistle or a  test tone.

Sawtooth or  Ramp  waves are what you get if you add up all the harmonics with amplitudes equal to 1/N where N is the number of the harmonic. Rich, bright. 

Square  waves are what you get by adding only the odd numbered harmonics with amplitudes equal to 1/n. Hollow, woody

So, Sawtooth  or Ramp  contain all harmonics whereas Square waves contain only odd numbered harmonics. 

Triangle waves are what you get if you add only the odd harmonics at decreasing intensity. Duller than square wave

Pulse  waves: A square wave is  a pulse wave with a 50% 'duty cycle' or a 50:50 ratio. If you think about the figure of a sine wave it is above the centre line for half of the time and below it for the other half . A square wave does the same. Synthesisers that generate pulse waves allow you to change the duty cycle. This in turn changes the harmonic content. A square wave with it's 50% duty cycle has a hollow, clarinet like tone, a pulse wave with, for example, a 5% duty cycle has a more nasal, double reed like tone.  Nasal, spikey.

White  noise is a hiss like you hear from a radio or TV when it is between stations. White noise is what you get when you add all audible frequencies together with equal energy. Its amplitude is constant throughout the audible frequency range. High-pitched hiss.

Pink  noise   is what you get if you add all frequencies with equal energy per octave. It sounds like white noise with the bass end enhanced. It is random noise that has the same power in each octave.   Smooth

top There are three main types of synthesis
  • Subtractive
  • Additive
  • Direct
Subtractive synthesis

Subtractive is where a rich source is filtered and enveloped.  A waveshape is chosen and has filter and amplitude envelopes applied to it.  A filter cuts down the number of harmonics and overtones in a waveform.  Each basic waveform has its own character.

  • sawtooth - quite bright - can be used for brass, strings, guitar and harmonica
  • square - hollow but bright - can be used for clarinet, brass, organ and harp-type sounds
  • pulse - quite thin - has a weak fundamental - resembles oboe, electric piano, sitar and vibraphone
  • sine and triangle - very pure - no harmonics in sine and only a few in triangle - used for recorder and tin whistle
  • white noise - used for laser sounds.
  • pink noise -used for rain
Additive synthesis

Additive is where waveforms (usually sine) are added together.  Each has its own amplitude (loudness) envelope.   Basic as well as complex waveshapes  can be created by adding the relevant harmonic frequencies at the correct loudnesses.  An envelope generator for each oscillator allows for tonal changes over time.

Direct synthesis

Direct  refers to  FM  (frequency modulation) and to synthesis taken from samples of sound.
With  sample-based  synthesis an actual digital recording - a sample - of a sound is made.  This may be processed further.

With FM  (patented by Yamaha) there are two or more oscillators which interact to create new frequencies, with the addition of the original and other harmonics. 


This gives you some idea of how
musicians create sounds electronically.

Student home