# rise and fall time

Hi all,

Does anyone know how we can extract the “rise and fall time” of a recorded sound, for example a piano note?

You mean past the obvious recording the note and then measuring the duration of each portion of the blue waves on the timeline? Most attacks are going to be consistent, but the decays are going to depend on the condition of the felts and whether or not you’re standing on the sustain pedal.

If you’re after what happens to the note with the sustain pedal down, the decay time is infinite. It’s asymptotic or something like that. You have to define at what level are you going to stop looking for it.

Koz

Quite an interesting question.
The attack is almost instantaneous - the hammer strikes the string and there you are.
Ok, reality is different. Certain harmonics join the Party rather late (e.g. a bell Sound has all its partials after about 2 s!).
The decay is made of several components - the decay of the fundamental, the decay of the partials (including “undertones”) and the room echo itself.
There’s the RT60 Standard which deals with the decay times of certain rooms. It gives you the time that a Sound Needs to decay to - 60 dB of its original gain.
I believe there’s a plug-in in the nyquist Forum that is dedicated to this Task.
But you can figure those values out for yourself.
The decay is seldom measured over 60 dB - normal equipement has a noise floor that is too high for that. But since the rate is constant, you can calculate it by simple multiplication. (20 dB in 0.5 s = 60 dB in 1.5 s).
The measurement starts at -5 dB from the highest Peak of your note and Ends about 15 dB over the noise floor (according to DIN).

Take a look at [u]ADSR[/u] (Attack, Decay, Sustain, Release).