Content-based recognition of musical instruments

Author

Fanelli, Anna Maria ; Caponetti, Laura ; Castellano, Giovanna ; Buscicchio, Cosimo Alessandro

Author_Institution

Dipt. di Informatica, Universita degli Studi di Bari, Italy

fYear

2004

fDate

18-21 Dec. 2004

Firstpage

361

Lastpage

364

Abstract

A method for content-based audio classification is presented. In particular we focus on identification of musical instruments sounds based on timbre classification, using a biologically plausible features extraction technique called cochleagram, and a new model of recurrent neural network called LSTM. Preliminary experiments are performed to compare various feature sets and neural network sizes. In particular two experiments are performed, using two different feature sets. The best classification rate obtained is 80%, averaged on 20 trials.

Keywords

audio databases; audio signal processing; content-based retrieval; feature extraction; musical instruments; recurrent neural nets; signal classification; audio classification; audio database; audio feature extraction; content-based recognition; musical instrument; recurrent neural network; Biological system modeling; Content based retrieval; Electronic mail; Feature extraction; Instruments; Music information retrieval; Neural networks; Pattern recognition; Recurrent neural networks; Timbre;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Information Technology, 2004. Proceedings of the Fourth IEEE International Symposium on

Print_ISBN

0-7803-8689-2

Type

conf

DOI

10.1109/ISSPIT.2004.1433794

Filename

1433794