feature computation

9 thoughts on “feature computation

    • Hi, do you mean the call of function ComputeFeature()? There is not much to write; let’s say you want to compute the Spectral Centroid of file example.wav with default settings:

      [x,fs] = wavread('example.wav');
      [v, t] = ComputeFeature ('SpectralCentroid', x, fs);

  1. Hi, thanks for sharing

    Sorry, I don’t have any programming experience. But does this mean that I can upload this code in MATLAB and get the features mentioned extracted from an uadio file. And if yes, how do upload the file?

    I am very eager to do ACA-analyses of approximately 100 audio files for a musicological purpose, but I am not sure how to get the features extracted – especially into a .csv file that I will be able to work with without having to program.

    Best, Jesper

    • This collection of Matlab scripts is more for showing how to code this rather than to use it as a non-programmer. You need Matlab, and you need at least basic understanding of Matlab’s script language. If I understand your case correctly, I think the best solution would be to use a tool such as Sonic Visualiser (http://www.sonicvisualiser.org/) with its selection of plugins.
      Hope that helps,

Leave a Reply

Your email address will not be published. Required fields are marked *