Nearest-neighbor automatic sound classification with a wordNet taxonomy

Mostra el registre complet Registre parcial de l'Ć­tem

  • dc.contributor.author Cano Vila, Pedro
  • dc.contributor.author Koppenberger, Markus
  • dc.contributor.author Le Groux, Sylvain
  • dc.contributor.author Ricard, Julien
  • dc.contributor.author Wack, Nicolas
  • dc.contributor.author Herrera Boyer, Perfecto, 1964-
  • dc.date.accessioned 2019-06-27T16:39:00Z
  • dc.date.available 2019-06-27T16:39:00Z
  • dc.date.issued 2005
  • dc.description.abstract Sound engineers need to access vast collections of sound efects for their film and video productions. Sound efects providers rely on text-retrieval techniques to offer their collections. Currently, annotation of audio content is done manually, which is an arduous task. Automatic annotation methods, normally fine-tuned to reduced domains such as musical instruments or reduced sound effects taxonomies, are not mature enough for labeling with great detail any possible sound. A general sound recognition tool would require first, a taxonomy that represents the world and, second, thousands of classifiers, each specialized in distinguishing little details. We report experimental results on a general sound annotator. To tackle the taxonomy definition problem we use WordNet, a semantic network that organizes real world knowledge. In order to overcome the need of a huge number of classifiers to distinguish many different sound classes, we use a nearest-neighbor classifier with a database of isolated sounds unambiguously linked to WordNet concepts. A 30% concept prediction is achieved on a database of over 50.000 sounds and over 1600 concepts.
  • dc.format.mimetype application/pdf
  • dc.identifier.citation Cano P, Koppenberger M, Le Groux S, Ricard J, Wack N, Herrera P. Nearest-neighbor automatic sound classification with a wordNet taxonomy. J Intell Inf Syst. 2005;24(2-3):99-111.
  • dc.identifier.issn 0925-9902
  • dc.identifier.uri http://hdl.handle.net/10230/41883
  • dc.language.iso eng
  • dc.publisher Springer
  • dc.relation.ispartof Journal of intelligent information systems. 2005;24(2-3):99-111.
  • dc.rights Ā© Springer The final publication is available at Springer via https://link.springer.com/content/pdf/10.1007%2Fs10844-005-0318-4.pdf
  • dc.rights.accessRights info:eu-repo/semantics/openAccess
  • dc.subject.keyword Audio identification
  • dc.subject.keyword WordNet
  • dc.subject.keyword Nearest-neighbor
  • dc.subject.keyword Everyday sound
  • dc.subject.keyword Knowledge management
  • dc.title Nearest-neighbor automatic sound classification with a wordNet taxonomy
  • dc.type info:eu-repo/semantics/article
  • dc.type.version info:eu-repo/semantics/acceptedVersion