Effectiveness Analysis of Features Used for Categorization of Music and Speech

Oezbayramoglu E., COŞAR A., Yazici A.

21st Signal Processing and Communications Applications Conference (SIU), CYPRUS, 24 - 26 Nisan 2013, (Tam Metin Bildiri) identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/siu.2013.6531339
  • Basıldığı Ülke: CYPRUS
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

Categorization and retrieval to multimedia for audio is a significant problem because of new technology has developed very rapidly. Content retrieval techniques are vital in order to gather information from categorized record. In literature Zero Crossing Rate (ZCR), Root Mean Square (RMS), Spectral Flux (SF), Spectral Centroid (SC) and Spectral Roll off (SR) features are used. Support Vector Machine (SVM) classifier is used for audio content analysis. Root Mean Square is more effective classifier feature than other features.