版本
Choosing the delay time parameter dictates the nature of artefacts that will heard using the algorithm. Short delay values can lead to amplitude modulations artefacts but do not suffer from distinct echo problem. On the other hand, using large delay values will no longer suffer from amplitude modulation problems and make the interval sound more in tune when compared to the dry signal on monophonic signals, but a distinct echo can be heard. Recommendation: A range between 10 ms to 100 ms is usually recommended for sounds that are just pitched (not harmonized). In those cases the echo is tolerable and high pitch resolution is not necessary since the original (dry) signal reference is not present.
The ideal delay time is dependent on audio content where signals with lots of transients (e.g. punches, drums, explosions,etc.) may benefit from smaller delay time settings while harmonic signals with more stable frequency content (e.g. voices, musical instruments, etc.) may work best with slightly larger delay time parameter values which provide more accurate frequency resolution.