lee-2009-audio Unsupervised feature learning for audio classification using convolutional deep belief networks Fork from the original code found in Lee's website http://web.eecs.umich.edu/~honglak/ Trying to port to Octave and allow the use of generic databases