The envelope represents the varying level of a sound wave over time and is broken down into four areas; attack, decay, sustain and release.
It is the envelope of a wave, in conjunction with the timbre, which helps establish the sound’s unique individual quality; the envelope has significant influence on how we interpret sound.
The envelope of a signal can be measured in four ways:
1. Attack – The portion of the envelope that represents the time taken for the amplitude to reach its maximum level. Essentially it is the initially build up of a sound.
Sound with obvious attack
2. Decay – The progressive reduction in amplitude of a sound over time. The decay phase starts as soon as the attack phase has reached its peak. In the decay phase, the signal level drops until it reaches the sustain level.
Sound with short decay (and long sustain)
3. Sustain – The period of time during which the sound is sustained before it begins to fade out. Many instruments do not contain a sustain phase.
Sound with long sustain (and no attach or delay)
4. Release – The final fade or reduction in amplitude over time.
Sound with obvious long release
On modern day synthesizers, the ADSR envelope is usually controlled via a set of control knobs:
Every envelope sound is different, for example, a percussion sound starts very suddenly, but then decays and releases quickly because no more energy is being applied to sustain the sound. A bowed string, on the other hand may build up with a slow attack, sustain for a short period and then release. Other sounds, such as an organ, can be sustained indefinitely by the player.
Most natural sounds decay the higher frequencies faster than the low frequency components because high frequency energy is dissipated more rapidly than low frequency energy.
The concept of hearing envelopes relies upon the root-mean-square values (RMS) values of amplitude and not peak to peak amplitudes - which is similar to the human perception of loudness. This means that high peaks in the signal will not necessarily make an instrument sound loud unless the amplitude is sustained for a period of time.
Short peaks tend to contribute to the character (timbre) of the sound rather than the loudness.