Abstract and Physical Synthesis#
Before we are diving into the topic of sound design via sound synthesis it is worth to mention that there are two distinct methods for sound synthesis:
abstract digital sound synthesis, and
physical modeling.
In abstract digitial sound synthesis we try to synthesis sound by focusing perception while in physical modeling we focus on the phyisical world (objects and their properties) and try to model them via, e.g., differential equations.
The advantage of physical modeling is that we have well interpretable parameters. Therefore, it is much easier for the composer to generate the sound she desires. In contrary, abstract synthesis such as additive synthesis, substractive synthesis or frequency modulation lack this interpretability. Furthermore, these methods often require thousand of parameters to generate interesting sounds which makes it hard to control them or to explore high-dimensional space of they span. However, the downside of physical modeling is that it requires a lot of computation. For example, in FM synthesis we achieve a rich sound by using only two oscillators, which is computational cheap. To generate the same richness using physical modeling requires much more operations.
The following techniques fall under the category of abstract digital sound synthesis:
Using abstract modeling we try to construct what we want directly and in physical modeling we try to create a model of ‘the world’ that is capable of generating what we want.