Segmentation of Singing, Speech and Instruments in Kunqu Audio Based on Zero-Crossing Rate
Gen-Fang Chen, Yadong Wu · 2019
Kunqu opera is one of the oldest Chinese traditional operas, Kunqu audio consists some fragment audio of singing, speech and instrument. This paper uses the zero-crossing rate characteristics of the audio to automatically segmentation Kunqu audio, and uses the Matlab tool to design the experiment. The experimental results show that the segmentation accuracy is 91.61%.