Music fingering is a cognitive process whose goal is to map each note of a music score to a fingering on some instrument. A fingering specifies the fingers of the hands that the player should use to play the notes. This problem arises for many instruments and it can be quite different from instrument to instrument; guitar fingering, for example, is different from piano fingering. Previous work focuses on specific instruments, in particular the guitar, and evolutionary algorithms have been used. In this paper, we propose a differential evolution (DE) algorithm designed for general music fingering (any kind of music instruments). The algorithm uses an Adaptive Neuro-Fuzzy Inference System (ANFIS) engine that learns the fingering from music already fingered. The algorithm follows the basic DE strategy but exploits also some customizations specific to the fingering problem. We have implemented the DE algorithm in Java and we have used the ANFIS network in Matlab. The two systems communicate by using the MatlabControl library. Several tests have been performed to evaluate its efficacy.

A Differential Evolution Algorithm Assisted by ANFIS for Music Fingering

DE PRISCO, Roberto;ZACCAGNINO, GIANLUCA;ZACCAGNINO, ROCCO
2012

Abstract

Music fingering is a cognitive process whose goal is to map each note of a music score to a fingering on some instrument. A fingering specifies the fingers of the hands that the player should use to play the notes. This problem arises for many instruments and it can be quite different from instrument to instrument; guitar fingering, for example, is different from piano fingering. Previous work focuses on specific instruments, in particular the guitar, and evolutionary algorithms have been used. In this paper, we propose a differential evolution (DE) algorithm designed for general music fingering (any kind of music instruments). The algorithm uses an Adaptive Neuro-Fuzzy Inference System (ANFIS) engine that learns the fingering from music already fingered. The algorithm follows the basic DE strategy but exploits also some customizations specific to the fingering problem. We have implemented the DE algorithm in Java and we have used the ANFIS network in Matlab. The two systems communicate by using the MatlabControl library. Several tests have been performed to evaluate its efficacy.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11386/3874702
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 4
  • ???jsp.display-item.citation.isi??? ND
social impact