Mirex 2012 Submission Audio Classification Using Sparse Feature Learning

We present a training/test framework for automatic audio annotation and ranking using learned feature representations. Commonly used audio features in audio classification, such as MFCC and chroma, have been developed based on acoustic knowledge. As an alternative, there is increasing interest in learning features from data using unsupervised learning algorithms. In this work, we apply sparse Restricted Boltzmann Machine to audio data, particularly focusing on learning high-dimensional sparse feature representation. Our evaluation results on two music genre datasets show that the learned feature representations achieve high accuracy.