| 摘要 |
PURPOSE:To reduce the quantity of samples of inverse transform and to shorten the analysis time by inversely transforming a part of the spectrum, which is obtained by Fourier transform of a sound signal in all bands, to obtain amplitude and frequency envelopes of each partial sound. CONSTITUTION:Musical sounds to be analyzed from a signal source 1 like a microphone are amplified to a proper level by a preamplifier 2 and are inputted to a fast Fourier transformer 4 through an AD converter 3. Musical sounds are converted to a frequency spectrum and are stored in a memory 5. A buffer 6 where the spectrum of partial sounds to be noticed is stored is prepared, and an inverse Fourier transformer 7 executes the inverse Fourier transform. This data is inputted to an amplitude envelope operating means 8 and a frequency envelope operating means 9 and is operated in accordance with respective operation formulas, and operation results are outputted from terminals 8A and 9B, thereby performing the operation at a high speed with a smaller number of samples. |
