Institutional Repository
Institutional Repository Institutional Repository Login  Institutional Repository  
  • UOC Library |
  •  |
  •  |

Home >
Academics >
IT, Multimedia and Telecommunications >
Telecommunication Technologies >
Bachelor thesis, research projects, etc. >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/8048
Title: Reconeixement automàtic d'instruments musicals
Authors: Castro Mayorgas, Juan Alberto
Director: Duxans Barrobés, Helenca
Other: Universitat Oberta de Catalunya
Keywords: automatic musical instruments recognition
musical instruments identification
pattern recognition
automatic indexing
musical signal processing
recognition of musical timbre
Issue Date: Jun-2011
Publisher: Universitat Oberta de Catalunya
Type: Bachelor Thesis
Abstract: The identification of musical instruments is one of the main areas of research to provide solutions to the need for automatic tools for indexing from the contents of recorded music. This paper studies the behavior of a set of relevant acoustic characteristics together with Gaussian Mixture Model, Support Vector Machines and k-Nearest Neighbor classification methods in the recognition of 18 musical instruments. Over 115,000 samples coming from TunedIt and another set of samples recorded by the author have been used for the experiments, reaching a success rate over 90 % in both cases. The study highlights that both the relevance of the acoustic parameters and the level of difficulty in identifying a particular instrument are related to the classifier.
Language: Catalan
URI: http://hdl.handle.net/10609/8048
Appears in Collections:Bachelor thesis, research projects, etc.


Add This:

RefWorks:

Ratings:
  0 (0 valuations)


Files in This Item:

File Description SizeFormat
jcastrom_TFC0611_memoria.pdfMemòria del projecte1.73 MBAdobe PDFPreview  Download
jcastrom_TFC0611_presentacio.ppsPresentació1.91 MBMicrosoft PowerpointPreview  Download
jcastrom_TFC0611_codi.zipCodi1.99 MBZipPreview  Download

Author names in Twitter


Author names in FriendFeed



Recommend this item

SFX Query

This item is licensed under a Creative Commons License
Creative Commons

Items in Repository e are protected by copyright, with all rights reserved, unless otherwise indicated.

 

The library replies
A product of the Universitat Oberta de Catalunya Virtual Library