Subtractive synthesis
The subtractive synthesis is a method of synthetic sound generation and is z. B. used in synthesizers .
Working principle
In subtractive synthesis, an oscillator (e.g. VCO ) generates a sound raw material, which is then processed by sound-changing modules ( filters , envelope generators, amplifier modules , etc.). The desired sound is achieved by filtering out or lowering the unwanted frequency components (= subtraction) from the oscillator's spectrum, which is mostly rich in overtones .
Oscillators
Most oscillators provide the following waveforms as raw material:
- Sawtooth , brilliant sound that contains all of the naturally occurring overtones. Is well suited for imitating string instruments ( audio sample here )
- Rectangular , clarinet-like to nasal sound that only contains the odd-numbered overtones. Well suited to imitate flute-like sounds ( audio sample here )
- Triangle, hollow sound that contains almost only the first five overtones. Well suited to thicken flat sounds ( audio sample here )
- Sinus , "round" sound that consists of only one fundamental tone. In practice, the sound of a tuning fork comes very close to a sine. Since a pure sine contains no overtones, the sound cannot be changed by linear filtering ( audio example here ).
The AI (Advanced Integrated) and AI Square synthesis from Korg , for example the Korg X5, is in principle a subtractive synthesis in which, instead of such raw material, correct sounds (waveforms) are used that are in a sample ROM.
The frequency spectrum generated by the oscillator can be influenced by other parameters. The pulse width modulation of square wave forms, the ring modulation of several oscillators, the frequency modulation of two oscillators and the synchronization of two oscillators, usually referred to as sync for short, are widespread here .
You can also change the symmetry of an oscillation in the course of the tone and thus, for example, let a shrill, asymmetrical sawtooth oscillation continuously transition into a gentle, symmetrical triangular oscillation, thereby creating a very dynamic timbre course (Synthesizer Workstation Pro).
filter
In the case of the filter, the low-pass filter ( audio example here ) is of particular importance as the most versatile variant . Many synthesizers only have a low-pass filter. Further forms are high-pass filters ( audio sample here ), band-pass filters (a combination of high-pass and low-pass, audio sample here ), notch filters ( audio sample here ) and comb filters .
The most important filter parameter is the cutoff frequency . The frequencies above the corner frequency are not completely removed, but are attenuated more and more as the distance from the corner frequency increases. How quickly the overtones become quieter depends on the so-called slope of the filter, which is specified in decibels per octave (dB / octave). A high slope leads to more brutal sound gradients, a low one to softer sound gradients. Most synthesizers have low-pass filters with a slope between 12 and 24 dB / octave.
In most synthesizers, the cutoff frequency of the filter and thus the sound shaping can be controlled manually using controls or automatically using modulators such as filter envelopes or LFOs . Filters also usually have a so-called resonance (Emphasis or Contour), which can also be controlled manually or automatically. The resonance raises the amplitudes of the frequencies in the range of the corner frequency, making the sound thinner and more nasal ( audio example here ).
The subtractive sound shaping is best audible with two-dimensional sounds, in which the filter corner frequency wanders through the sound spectrum while the tones are held ( audio example here ).
amplifier
After the filter section, the signal is sent to an amplifier, the VCA (Voltage Controlled Amplifier). This is also decisive for the sound quality and sound characteristics of the signal. If the sound was synthesized solely from the oscillator, filter and amplifier, a standing tone would result. This is changed by using so-called envelopes of the VCA.
Other occurrences
The human voice also produces its sounds using a kind of subtractive synthesis. The larynx produces a sound rich in overtones, which is then filtered through the mouth . The "U" has a lower corner frequency. Almost all overtones are cut off here, so that almost only the pure sine fundamental remains. The "A" is a sound rich in overtones that is only slightly filtered.
See also
- Additive synthesis | FM synthesis | Granular Synthesis | Wavetable Synthesis | Sampling synthesis
- Sampling | Physical modeling
literature
- Synthesizer Workstation Pro. Franzis, Poing near Munich 2010, ISBN 978-3-645-70094-8 .
Individual evidence
- ↑ Synthesizer / synthese-formen ( Memento from June 26, 2009 in the Internet Archive ), please note paragraphs 2.12 and 2.13 here
- ↑ Subtractive sound synthesis - The basic building blocks ( Memento from April 23, 2019 in the Internet Archive )
