Karplus–Strong string synthesis

The Karplus-Strong Algorithm to a process of the synthesis sound, which is particularly suitable to simulate the sounds of stringed instruments. The algorithm was introduced in 1983 in the Computer Music Journal by Kevin Karplus and Alex Strong. A U.S. patent number 4,649,783 was granted in 1987. The algorithm is based on a cycle, circulate in the noise samples and thereby pass through a filter. In this circuit, a random noise is initially fed, which is modified in each pass, and output again from the filter.

Operation

First of p samples with random noises are introduced into the circuit, which fill a memory ( FIFO) which can hold samples p. Thereafter, the entry is terminated and the first sample leaves the memory is output through the filter and is charged again into the memory is now in its modified version. It is all done at a rate of R samples per second, which leave the store and go through the cycle.

Due to a psychoacoustic phenomenon, the " Periodicity Pitch" (periodic pitch) one perceives a tone of height R / p Hz. The timbre is determined by the selected input noise. To simulate the decay characteristic of a plucked string, a low-pass filter is used as filter. High-frequency components of the noise thereby lose energy more rapidly than the low frequency.

Comments

The input samples are particularly suitable noise with dense range, such as a white noise. If you want to simulate different strengths in the string stopper, one should allow the samples before the first edition a few passes circulate without issue. This causes a more realistic effect than simply adjust the volume before output.