Spectral synthesis and string synthesis blog

Spectral modeling synthesis is a set of techniques and software implementations for the analysis, synthesis, and transformation of music based on a sinusoidal plus residual model. These techniques can be used for a variety of things related to music, such as synthesis, sound source separation, music perception, performance analysis, and more. This model was developed by X. Serra in 1989.

Karplus-Strong string synthesis is a type of synthesis that loops a short waveform through a filtered delay line to create a noise that sounds like a plucked string or some form of percussion. This type of synthesis is known to be a subtractive synthesis technique based on a feedback loop similar to a comb filter for z-transform analysis. This type of synthesis is also viewed as the simplest of a class of wavetable-modification algorithms since the delay line acts to store one period of the signal. String synthesis was developed by Alexander Strong, and Devin Karplus did the first analysis of how it worked, hence it being named Karplus-Strong string synthesis. The first song that used this type of synthesis was a song called “May All Your Children Be Acrobats” written in 1981 by David A. Jaffe.