Peakamp~

This patch shows different ways to receive and generate numerical values using MaxMSP. 

To detect when an audio event (such as a musical note) occurs, one straightforward method is to test whether the peak amplitude of the audio signal surpasses an established threshold that's slightly above the level of the ambient noise floor. The peakamp~ object periodically reports the greatest absolute value of amplitude that has occurred during the specified time interval. For quick response, that interval should usually be every 10 to 30 milliseconds.

This patch is very similar to the "Simple envelope follower", with the difference that in this patch the line~ object's time of interpolation to get to a detected peak value can be different from the peakamp~ object's interval of evaluation. If, for example, the peakamp~ interval is 10 ms and the line~ ramp time is 2.5 ms, the envelope follower latency will range from only 2.5 to 12.5 ms.

An "envelope follower" provides a smoothed global representation of the extreme amplitudes of a signal. It can be as rough or as detailed as you want it to be, depending on how much of the original signal you disregard in the evaluation of the peaks. Because sound signals tend to vary in both positive and negative directions around a central 0 value, it's best to evaluate the absolute values of the samples, so that peaks in the negative direction are easily compared to peaks in the positive direction.

A "ducker" is a system that turns a signal down to 0 when it's below a given threshold. This is also frequently called a "noise gate" in audio engineering, because it acts as a doorway that closes out unwanted low-level ambient noise and lets through only the louder, more desired signal. It's useful for suppressing unwanted low-level audio, such as in a cell phone transmission when the user is not talking, or, more to the point for musical purposes, as in a microphone signal when the musician is not playing.

Here’s a way to create a four-way audio crossfader.

The example demonstrates two methods. This first one uses smoothing to create a ramp on and off instead of an instantaneous switch on and off. This second one waits a certain amount of time before it begins the release ramp, and if the input signal goes back above the threshold during that time it cancels the release (the default amplitude threshold being -40 dB).

The peakamp~ object can be used to take the amplitude of one signal and have it control the amplitude of another. Peakamp~ takes a low-frequency sample of the peak amplitude of the signal, which is then smoothed by the line~ object. This gives an intuitively much more accurate representation of the audio signal one is envelope-following than does smoothing the audio signal itself, which tends to smooth out — and thus lose — all the more interesting characteristics of the signal. Here’s an example that lets you compare.