Loading...
Logo
Processing Forum
Contributed Library Questions
sabine
Sabine

analyse sound

in Contributed Library Questions  •  1 year ago  
Hey all,

I want to analyse about 5 minutes of sound and I want
to return variables per 1 or 2 second that collect the
averages of the following.

the average of the period of the high, middle and low.
I has to be returned within 0 and 255.
Also the average of the volume of the period.

realy hope you guys can help me out!

Replies(1)

Christian Schulze
Christian..

Re: analyse sound

1 year ago
seems like this is a whole program you're asking for. have a look at the minim library. you can get the high, middle and low frequencies with a fast fourier transform (fft). try the following sketch as a start:
Copy code
  1. import ddf.minim.*;
  2. import ddf.minim.analysis.*;

  4. Minim minim;
  5. AudioPlayer sound;
  6. FFT fft;

  8. void setup() {
  9.   size(640, 360);
  10.  
  11.   minim = new Minim(this);
  12.   sound = minim.loadFile("yoursoundfile.mp3");
  13.   sound.play();
  14.   fft = new FFT(sound.bufferSize(), sound.sampleRate());
  15. }

  17. void draw() {
  18.   background(255);
  19.   fft.forward(sound.mix);
  20.   for (int i = 0; i < fft.specSize(); i++) {
  21.     int tempBand = (int) fft.getBand(i);
  22.     line(i, height, i, height - tempBand);
  23.    
  24.     // better look, but cut off in high frequencies
  25.     //rect(i * 4, height, 4, -tempBand);
  26.   }
  27. }

  29. void stop() {
  30.   sound.close();
  31.   minim.stop();
  32.   super.stop();
  33. }
your sound file needs to be in a folder called "data" within your sketch folder (/yoursketch/data/yoursoundfile.mp3).

fft.getBand(i) returns the value of the frequency at position i. i therefor has to be within the range of fft.specSize(). for the averages of low, mid and high frequencies you need to divide the specSize into three ranges. within these three ranges you can calculate the averages of the frequency values.

here's the official manual on the library website:
http://code.compartmental.net/tools/minim/manual-fft/

and here's the javadoc:
http://code.compartmental.net/minim/javadoc/index.html?ddf/minim/analysis/FFT.html


Cheers,
Chris
Top Reply

Statistics

  • 1 Replies
  • 506  Views
  • 1 Followers

Tags

No tags available for this topic.

Actions

Related Posts