Hi

What I want to create is a visual analysis of short sound files as images eg. of different people saying the same word with different accents or in different languages, for comparison. I want to print this image or pull it into Illustrator and play with it, so I want to export it as a PDF.

So far I think the best way to do this is use Minum to get the frequency of the sound file and then draw a line having the frequency determine the y-position, moving along the x-axis over time to draw a basic graph and then save a PDF file. This is probably pretty basic, but I've only been playing with Processing for a day and I just want some help with direction, as I've been running down a lot of dead ends.

I've been looking at some examples that render a frequency analyis in real time (eg. http://www.ee.columbia.edu/~dpwe/e4896/code/PlaySpectrogram_thresh.pde), but what I want is just to generate a static image of the entire sound file, and preferably using a simple line so I have a nice clean image.

a.) How do I output a frequency analysis, not in real time, but of the whole sound file? Is there a way to draw this in real time and then save a PDF? When I try this I get a conflict of modes

b.) should I actually be creating a sketch that creates a text file of the sound and then make another sketch that uses data from that file to draw a graph / line to avoid the mode conflict, or is this just a really long way round of doing things?

thanks :) any suggestions would be greatly appreciated