论文信息 - Content-Based Audio Classification and Retrieval Using SVM Learning

Content-Based Audio Classification and Retrieval Using SVM Learning

In this paper, a support vector machines (SVMs) based method is proposed for content-based audio classification and retrieval. Given a feature set, which in this work is composed of perceptual and cepstral feature, optimal class boundaries between classes are learned from training data by using SVMs. Matches are ranked by using distances from boundaries. Experiments are presented to compare various classification methods and feature sets.

Guodong Guo | Stan Z. Li | Stan Z. Li | G. Guo

[1] Stan Z. Li,et al. Face recognition using the nearest feature line method , 1999, IEEE Trans. Neural Networks.

[2] B. Feiten,et al. Automatic indexing of a sound database using self-organizing neural nets , 1994 .

[3] W. Andrew Schloss,et al. Toward an Intelligent Editor of Digital Audio: Signal Processing Methods , 1982 .

[4] Vladimir Vapnik,et al. Statistical learning theory , 1998 .

[5] Douglas Keislar,et al. Content-Based Classification, Search, and Retrieval of Audio , 1996, IEEE Multim..

[6] Wolfgang Effelsberg,et al. Automatic audio content analysis , 1997, MULTIMEDIA '96.

[7] Stan Z. Li,et al. Content-based Classification and Retrieval of Audio Using the Nearest Feature Line Method , 2000 .

[8] Tamas Ungvary,et al. Organization of Sounds with Neural Nets , 1991, ICMC.

[9] Tsuhan Chen,et al. Audio feature extraction and analysis for scene classification , 1997, Proceedings of First Signal Processing Society Workshop on Multimedia Signal Processing.

[10] Jonathan Foote,et al. Content-based retrieval of music and audio , 1997, Other Conferences.

[11] B. S. Manjunath,et al. Texture Features for Browsing and Retrieval of Image Data , 1996, IEEE Trans. Pattern Anal. Mach. Intell..