Musical notes recognition using artificial neural networks

A system to convert the image of such a score to a sound file is useful in many areas such as: control, robotics, computer vision. The purpose of this paper is to develop a method to obtaining a direct conversion from a musical score to a sound file. The paper presents the design methodology of an a...

Full description

Saved in:
Bibliographic Details
Published inAnnals of DAAAM & proceedings Vol. 20; no. 1; p. 1159
Main Authors Moise, Adrian, Constantin, Adrian, Bucur, Gabriela
Format Journal Article
LanguageEnglish
Published DAAAM International Vienna 01.01.2009
Subjects
Analysis
Musical notation
Neural networks
Austria
Online AccessGet full text
ISSN1726-9679
1726-9679

Cover

More Information
Summary:A system to convert the image of such a score to a sound file is useful in many areas such as: control, robotics, computer vision. The purpose of this paper is to develop a method to obtaining a direct conversion from a musical score to a sound file. The paper presents the design methodology of an artificial neural network (ANN) used to recognize musical notes. First, the image of the musical score is captured. One of the main contributions of the authors is identifying and extracting the features of the musical score and exporting them to the ANN. Then, the input/output codes are described, the network is trained and the results of different training methods with different neurons are compared. Key words: musical note, artificial neural network, object features, training algorithm
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:1726-9679
1726-9679