About sound synthesis methods. Sound Synthesis Methods
A single-voice or monophonic synthesizer is a synthesizer that can play only one note at a time, that is, you will not play a chord on such a synthesizer. Rather, you take it, but only one of the notes of the chord will sound. As a rule, the notes will alternate, for example, you have several chords in different keys, from which a melody or some kind of pad is composed.
A polyphonic or polyphonic synthesizer is capable of playing several notes at the same time (no more than the number of its voices) at any given time. I explain: if, for example, you have an 8-voice synthesizer, then at the same time it can play no more than eight notes of one tone. A multi-timbral synthesizer is capable of playing several notes at once with different tones.
Additive synthesis method
The very first synthesis method was the additive method (English add – add). This method is based on the addition of waves of several generators. The method is based on the Fourier theorem; it consists in the fact that any periodic oscillation can be represented as the sum of sinusoidal oscillations of various frequencies and amplitudes.
This method is used in all existing synthesizers with more than one wave generator. If sinusoidal oscillations with multiple frequencies (differing by an integer number of times) are used as the initial ones and the amplitudes can be different, then this type of synthesis is called harmonic (this term corresponds to a harmonic scale in which the frequencies of the same notes of neighboring octaves differ by half).
Another type of additive synthesis is register synthesis. In this case, oscillations of more complex shapes, for example, triangular or straight, are used as the initial waves.
Subtractive synthesis method
If literally translated from English, you get a “subtractive method” (subtract – subtract). This method consists in the fact that a new timbre is obtained by changing (subtracting) the components in the spectrum of the initial oscillation. This process takes place in two stages. First, the main oscillations are formed, the main requirement for the initial oscillation is to ensure that initially it (the oscillation) has the most developed timbre, that is, a lot of spectral components. For example, the timbre of a sine wave is simple and inexpressive, and the spectrum of, say, a sawtooth wave, is already brighter (I hope it’s clear what I’m talking about).
At the second stage, using frequency filters from the initial oscillation, frequency components corresponding to the simulated instrument are isolated. The main synthesis tools for this method are controlled filters: resonant (band-pass) – with a variable position (main response frequency) and bandwidth (band) and a low-pass filter (low-pass filter) – with a variable cutoff frequency.
The advantage of this method is its fairly simple implementation and a wide range of synthesized sounds. The main disadvantage is that when synthesizing sounds with a complex spectrum, you have to use a large number of filters.
As a rule, both additive and subtractive methods coexist peacefully and complement each other.
It is one of the latest achievements in the field of sound synthesis. To implement it, much more powerful microprocessors are needed than with subtractive or additive methods. A synthesizer of this type was created by R. Mug in 1964. Its basis was a voltage-controlled generator capable of generating rectangular, sawtooth and sinusoidal signals. Various connection options for such generators, as well as the subsequent addition of their output signals, made it possible to obtain a huge number of new “electronic” sounds.
This synthesis method is called FM additive synthesis. The method is based on Frequency Modulation. The scientific definition is something like this: a change in the frequency of a signal according to the law of some control voltage.
In FM synthesis, sound with the necessary timbre is produced on the basis of several sound frequency generators with their mutual modulation. The combination of the generator and the circuit that controls this generator is called the operator. In a combination of two operators, the one that directly reproduces the sound is called the carrier, and the one that modulates the wave is called the modulator.
The connection scheme of two or more operators is called an algorithm. The connection diagram of the operators, and the settings of each operator (frequency, amplitude and the law of their change in time) determine the sound timbre.
In a chain of series-connected operators, the last of them is a carrier (sound source), and the rest are modulators. The signal from the output of the operator can go to the input of its modulation in frequency.