Block-Level Audio Features for Music Genre Classification.

Beschreibung

While frame-level audio features, e.g. MFCCs, in combi-nation with the bag-of-frames approach have widely and successfully been used, we use a block processing framework in our submission. In general block-level fea-tures have the advantage that they can capture more tem-poral information than BOF approaches can. We intro-duce two novel spectral patterns, closely related to the spectrum histogram and propose a modified version of the well-known fluctuation patterns. Based on these pat-terns we train a support vector machine to classify songs into different categories.

Zeitraum	28 Okt. 2009
Ereignistitel	10th International Conference on Music Information Retrieval (ISMIR 2009)
Veranstaltungstyp	Konferenz
Ort	Kobe, JapanAuf Karte anzeigen