Incorporating scale information with cepstral features: experiments on musical instrument recognition

Marcela Morvidone, Bob L. Sturm, Laurent Daudet

Research output: Contribution to journalJournal articleResearchpeer-review

10 Citations (Scopus)
152 Downloads (Pure)

Abstract

We present two sets of novel features that combine multiscale representations of signals with the compact timbral description of Mel-frequency cepstral coefficients (MFCCs). We define one set of features, OverCs, from overcomplete transforms at multiple scales. We define the second set of features, SparCs, from a signal model found by sparse approximation. We compare the descriptiveness of our features against that of MFCCs by performing two simple tasks: pairwise musical instrument discrimination, and musical instrument classification. Our tests show that both OverCs and SparCs improve the characterization of the global timbre and local stationarity of an audio signal than do mean MFCCs with respect to these tasks.
Original languageEnglish
JournalPattern Recognition Letters
Volume31
Pages (from-to)1489-1497
Number of pages9
ISSN0167-8655
DOIs
Publication statusPublished - 2010
Externally publishedYes

Bibliographical note

Part of this work was carried out while Dr. Sturm was a postdoctoral fellow at Université Pierre et Marie Curie, Paris 7; the submission was corrected and finalized while Dr. Sturm was at ADMT, AAUK.

Fingerprint

Dive into the research topics of 'Incorporating scale information with cepstral features: experiments on musical instrument recognition'. Together they form a unique fingerprint.

Cite this