A sound features extracting unit receives a sound signal of a music piece to extract sound features. A peak detecting unit detects a peak time when a cost function calculated by a cost function calculating unit indicates a maximal value, and a peak value thereof. A selecting unit selects a time property indicating sound critical points from the peak time and the peak value. This method allows the user to automatically retrieve sound critical points without having to listen to the sound in advance.

 
Web www.patentalert.com

< Speech input system, speech portal server, and speech input terminal

< Sound actuated display device incorporating vibratory-rotary motion converter

> Cascaded hidden Markov model for meta-state estimation

> Music player applicable to portable telephone terminal

~ 00222