Scalability in Content-Based Navigation of Sound Databases - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2009

Scalability in Content-Based Navigation of Sound Databases

Résumé

The article presents methods for sound search in large effects or instrument sound databases by interactive content-based navigation in a space of descriptors and categories, based on the principle of real-time corpus-based concatenative synthesis. We focus on three algorithms: fast similarity-based search by a kD-Tree in the high-dimensional descriptor space, a mass--spring model with added repulsion for layout, and efficient dimensionality reduction for visualisation by hybrid multi-dimensional scaling based on these. Special attention is given to scalability to very large databases by performance evaluations and measurements. The algorithms are implemented and tested as C-libraries and Max/MSP externals within a prototype sound exploration application.
Fichier principal
Vignette du fichier
index.pdf (237.68 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01161240 , version 1 (08-06-2015)

Identifiants

  • HAL Id : hal-01161240 , version 1

Citer

Diemo Schwarz, Norbert Schnell, Sébastien Gulluni. Scalability in Content-Based Navigation of Sound Databases. International Computer Music Conference (ICMC), Aug 2009, Montreal, Canada. pp.1-1. ⟨hal-01161240⟩
120 Consultations
79 Téléchargements

Partager

Gmail Facebook X LinkedIn More