Conference paper
Decision time horizon for music genre classification using short time features
In this paper music genre classification has been explored with special emphasis on the decision time horizon and ranking of tapped-delay-line short-time features. Late information fusion as e.g. majority voting is compared with techniques of early information fusion such as dynamic PCA (DPCA). The most frequently suggested features in the literature were employed including mel-frequency cepstral coefficients (MFCC), linear prediction coefficients (LPC), zero-crossing rate (ZCR), and MPEG-7 features.
To rank the importance of the short time features consensus sensitivity analysis is applied. A Gaussian classifier (GC) with full covariance structure and a linear neural network (NN) classifier are used.
| Language: | English |
|---|---|
| Publisher: | IEEE |
| Year: | 2004 |
| Pages: | 1293-1296 |
| Proceedings: | 2004 12th European Signal Processing Conference (EUSIPCO) |
| ISBN: | 3200001658 and 9783200001657 |
| Types: | Conference paper |
| ORCIDs: | Larsen, Jan |
decision time horizon dynamic PCA feature ranking majority voting music genre classification
Abstracts Feature extraction Gaussian classifier LPC MFCC MPEG-7 feature Mel frequency cepstral coefficient Rocks Stacking TV ZCR audio signal processing cepstral analysis classification consensus sensitivity analysis delay lines information fusion linear neural network classifier linear prediction coefficient mel frequency cepstral coefficient music neural nets principal component analysis sensitivity analysis sensor fusion tapped delay-line short time feature zero-crossing rate
