From: A Decision-Tree-Based Algorithm for Speech/Music Classification and Segmentation
Duration (seconds)
correct %
Music
78028
97.8%
Speech
42430
99.4%