The present work contributes to the field of generalized sound classification. We extensively examine the performance of the next three feature sets: a) MPEG-7 Audio Spectrum Projection, b) MFCC (using an alternative method for their extraction) and c) a group derived utilizing critical band based wavelet packets. Subsequently three types of tem poral feature integration strategies are applied on the extracted instant values: a) short-term statistics, b) spectral moments and c) two autoregressive functions. During the experimental phase, we organize ten sound classes using professional sound effects collections of high quality. The density of each category is approximated with left-right hidden Markov models. Comparable results with respect to all the feature sets as well as integration methods are provided, which demonstrate the superiority of the short-term statistics method.

Sound classification based on temporal feature integration / S. Ntalampiras, I. Potamitis, N. Fakotakis - In: 2010 4th International Symposium on Communications, Control and Signal Processing (ISCCSP)[s.l] : IEEE, 2010. - ISBN 9781424462858. - pp. 1-4 (( Intervento presentato al 4. convegno International Symposium on Communications, Control, and Signal Processing tenutosi a Limassol nel 2010 [10.1109/ISCCSP.2010.5463315].

Sound classification based on temporal feature integration

S. Ntalampiras;
2010

Abstract

The present work contributes to the field of generalized sound classification. We extensively examine the performance of the next three feature sets: a) MPEG-7 Audio Spectrum Projection, b) MFCC (using an alternative method for their extraction) and c) a group derived utilizing critical band based wavelet packets. Subsequently three types of tem poral feature integration strategies are applied on the extracted instant values: a) short-term statistics, b) spectral moments and c) two autoregressive functions. During the experimental phase, we organize ten sound classes using professional sound effects collections of high quality. The density of each category is approximated with left-right hidden Markov models. Comparable results with respect to all the feature sets as well as integration methods are provided, which demonstrate the superiority of the short-term statistics method.
Computer Networks and Communications; Signal Processing; Electrical and Electronic Engineering
Settore INF/01 - Informatica
2010
Cyprus University of Technology
University of Cyprus
Book Part (author)
File in questo prodotto:
File Dimensione Formato  
05463315.pdf

accesso riservato

Tipologia: Publisher's version/PDF
Dimensione 2.84 MB
Formato Adobe PDF
2.84 MB Adobe PDF   Visualizza/Apri   Richiedi una copia
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2434/615052
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 2
  • ???jsp.display-item.citation.isi??? ND
social impact