| 摘要 |
The invention relates to a method (1) for restoring partials of a sound signal between a peak Pi and a peak Pi+N, of which the frequencies O and phases are known. The inventive method (1) comprises the following steps: the frequency O of each of the missing peaks Pi+1 to Pi+N-1 of a partial is estimated (2); the executed phase Pi+N is calculated (3) from peak to peak, from the phase of the peak Pi to that of the peak Pi+N, for all of the frequencies O previously estimated; the error of phase err between the executed phase and the known phase at the same peak Pi+N is calculated (4); and each executed phase is corrected (5) by a value according to the error of phase err.
|
