| 摘要 |
An electronic musical instrument includes a first memory in which audio samples of lower frequency components of an aperiodic waveform are stored and a second memory in which audio samples of a higher frequency components of the waveform are stored. Digital samples stored in a first portion of the second memory represent a rapidly rising portion of the higher frequency waveform and those stored in a second portion of the memory represent a rapidly declining portion of the higher frequency waveform whose amplitude and spectral energy distribution profiles are preferably equalized. The first memory is addressed throughout in forward scan to generate a first output waveform. The second memory is addressed in an initial forward scan throughout its first and second portions and the direction of scan is reversed at the end of the second portion to recyclically address it in rearward and forward directions to generate a second output waveform, which is combined with the first output waveform. A monotonically declining envelope is preferably impressed upon the second output waveform to reconstruct the original declining amplitude and spectral energy distribution profile. |
