Generating guitar tablature with LHF notation via DGA and ANN

Abstract

This paper describes a system for converting music to guitar tablature. At run time, the system employs a distributed genetic algorithm (DGA) to create tablature and an artificial neural network to assign fingers to each note. Three additional genetic algorithms are used to optimize the fitness function of the DGA, the operating parameters of the DGA, and the learning environment of the Neural Network. These steps are taken in the hope of maximizing the consistency of our algorithm with human experts. The results have been encouraging. © Springer-Verlag Berlin Heidelberg 2006.