The original Karplus-Strong algorithm is based on wavetable synthesis. In wavetable synthesis, a buffer is fill with samples that are then read by a pointer that circulate from the beginning of the table to the end and back to the beginning, thus creating a periodic sound. The wavetable can be fill with any kind of wave. Triangular, sinusoidal and square wave are normally used but even a random signal like noise will sound periodic. The initial conditions of the wavetable determine the resulting timbre and its length determines the periodicity (i.e. the pitch) of the sound.

Figure 1 - Wavetable synthesis

Wavetable synthesis is very simple but unfortunately also very dull because the sound doesn't present any variation over time. Traditional instruments, on the other side, create sounds that vary a lot. Most synthesis techniques remedy to this situation by modifying the sound after it is read from the wavetable. The Karplus-Strong algorithm differs from those techniques by the fact that it directly modifies the wavetable at each of its iteration. The wavetable in the Karplus-Strong algorithm can therefore be seen as a delay-line.