EUSIPCO'2002 banner

Paper data
A Speech/Music Discriminator using RMS and Zero-crossings

Panagiotakis Costas, University of Crete
Tziritas Georgios, University of Crete

Page numbers in the proceedings:
Volume III pp 459-462

Content based Audio and Video Indexing (2/2)

Paper abstract
An audio segmentation method and a speech/music classifier are proposed. The characteristics used are considerably reduced. Segmentation is based on mean signal amplitude distribution, whereas classification utilizes an additional characteristic related to the mean frequency. The segmentation and classification algorithms were benchmarked on a large dataset, with correct segmentation about 97% of the time and correct classification about 95%.

A PDF version is available here