help with visualising spectogram

help with visualising spectogram

in Core Library Questions • 1 year ago

Hi,

I want a sketch which can analyse the dominant frequency band of a sound. So for example, with a kick drum it would analyse all the frequencies and tell me the low frequencies are the loudest in the spectrum. At the moment I have got a spectogram, but it shows all frequencies. How can I adapt this so it tells me which band has the most gain in it please?

import ddf.minim.analysis.*;
import ddf.minim.*;
Minim minim;
AudioPlayer jingle;
FFT fftLin;
int xPos = 0;
int hueLimit = 160;
void setup()
{
// create a screen width=512, height=256
size(512, 256);
// create a minim audio application object
minim = new Minim(this);
// create a file object (can be a MP3 file or a WAV file)
//jingle = minim.loadFile("jingle.mp3", 512);
jingle = minim.loadFile("Drum_Loop_01.wav", 512);
// loop the file
jingle.loop();
// create an FFT object that has a time-domain buffer the same size as jingle's sample buffer
// note that this needs to be a power of two
// and that it means the size of the spectrum will be 1024.
// see the online tutorial for more info.
fftLin = new FFT(jingle.bufferSize(), jingle.sampleRate());
// calculate the averages by grouping frequency bands linearly. use 30 averages.
fftLin.linAverages(30);
}
void draw()
{
// perform a forward FFT on the samples in jingle's mix buffer
// note that if jingle were a MONO file, this would be the same as using jingle.left or jingle.right
fftLin.forward(jingle.mix);
for(int i = 0; i < fftLin.specSize(); i++)
{
float hueValue = hueLimit - map((fftLin.getBand( fftLin.specSize()-i)*32), 0, 256, 0, hueLimit);
// colorMode is HSB, the range for hue is 256, for saturation is 100, brightness is 100.
colorMode(HSB, 256, 100, 100);
// color for stroke is specified as hue, saturation, brightness.
stroke(int(hueValue), 100, 80);
// find the distance to the edge of an eclipse defined by i from the center along the angle xPos; the numbers are basically //guessing and checking. I got the math from http://warpycode.wordpress.com/2011/01/21/calculating-the-distance-to-the-//edge-of-an-ellipse/
float elipse=i*.5*sqrt( 1/( sq(sin(xPos)/(.8)) + sq(cos(xPos)/(1.25))) );
//convert the distance and angle from polar to cartesian form
point(width/2+elipse*cos(xPos),height/2+elipse*sin(xPos));
}
// increment the x position
xPos = xPos + 1;
// wrap around at the screen width
if (xPos > width) {
xPos = 0;
}
}
void stop()
{
// always close Minim audio classes when you are done with them
jingle.close();
// always stop Minim before exiting
minim.stop();
//super.stop();
}

Replies(2)

Jonny Stu..

Re: help with visualising spectogram

1 year ago

You'll get both by decreasing the buffer size e.g.:
jingle = minim.loadFile("Drum_Loop_01.wav", 128);

Some of that code doesn't make much sense. sin() and cos() take a value in radians so when you feed then with an xPos that's being incremented by 1 every time you're covering an entire circle in 6 frames (2 * pi being one circle). There's also no reason to reset xPos when it's greater than width. I'd have thought it would make more sense for xPos to be a float and increase it by around 0.2 each frame but maybe how it is is how you want it.

joshmyc

Re: help with visualising spectogram

1 year ago

o.k. thanks johnny. I'm still unclear on how to tackle the code so that I can use fft to find out the loudest frequency band?

Top Reply

Response title

Attachments

New to this Portal?