Automatic string detection for bass guitar and electric guitar

In this paper, we present a feature-based approach to automatically estimate the string number in recordings of the bass guitar and the electric guitar. We perform different experiments to evaluate the classification performance on isolated note recordings. First, we analyze how factors such as the instrument, the playing style, and the pick-up settings affect the performance of the classification system. Second, we investigate, how the classification performance can be improved by rejecting implausible classifications as well as aggregating the classification results over multiple adjacent time frames. The best results we obtained are f-measure values of F = .93 for the bass guitar (4 classes) and F = .90 for the electric guitar (6 classes).