A Graphical Representation and Dissimilarity Measure for Basic Everyday Sound Events - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Article Dans Une Revue IEEE Transactions on Audio, Speech and Language Processing Année : 2012

A Graphical Representation and Dissimilarity Measure for Basic Everyday Sound Events

Anniés Robert
  • Fonction : Auteur
  • PersonId : 943752
Wahlen Elio
  • Fonction : Auteur
  • PersonId : 943753
Purwins Hendrik
  • Fonction : Auteur
  • PersonId : 943754

Résumé

Studies of Gaver (W. W. Gaver, "How do we hear in the world? Explorations in ecological acoustics," Ecological Psy- chology, 1993) revealed that humans categorize everyday sounds considering the processes that have generated them: He defined these categories in a taxonomy according to the aggregate states of the involved materials (solid, liquid, gas) and the physical nature of the sound generating interaction such as deformation, friction, etc., for solids. We exemplified this taxonomy in an everyday sound database that contains recordings of basic isolated sound events of these categories. We used a sparse method to represent and to visu- alize these sound events. This representation relies on a sparse de- composition of sounds into atomic filter functions in the time-fre- quency domain. The filter functions maximally correlated with a given sound are selected automatically to perform the decompo- sition. The obtained sparse point pattern depicts the skeleton of the given sound. The visualization of these point patterns revealed that acoustically similar sounds have similar point patterns. To de- tect these similarities, we defined a novel dissimilarity function by considering these point patterns as 3-D point graphs and applied a graph matching algorithm, which assigns the points of one sound to the points of the other sound. This novel dissimilarity measure is used in combination with a kernel machine for the classification experiments, yielding an average accuracy of 95% in one versus one discrimination tasks.
Fichier non déposé

Dates et versions

hal-00684620 , version 1 (02-04-2012)

Identifiants

  • HAL Id : hal-00684620 , version 1

Citer

Kamil Adiloglu, Anniés Robert, Wahlen Elio, Purwins Hendrik, Obermayer Klaus. A Graphical Representation and Dissimilarity Measure for Basic Everyday Sound Events. IEEE Transactions on Audio, Speech and Language Processing, 2012, 20 (5), pp.1542-1552. ⟨hal-00684620⟩
162 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More