RESEARCH ARTICLE Improving Music Genre Classification Using Automatically Indu ced Harmony Rules
Emmanouil Benetos, Matthias Mauch, Simon Dixon · 2010
We present a new genre classification framework using both low -level signal-based features and high-level harmony features. A state of-the-art statistical genre clas sifier based on timbral features is extended using a first-order random forest containing for each genre rules d erived from harmony or chord sequences. This random forest has been automatically induced, using the first -order logic induction algorithm TILDE, from a dataset, in which for each chord the degree and chord category are identified, and covering classical, jazz and pop genre classes. The audio descriptor-based genre classifier contains 206 features, covering spectral, temporal, energy, and pitch characteristics of the audio sig nal. The fusion of the harmony-based classifier with the extracted feature vectors is tested on three-genre subsets of the GTZAN and ISMIR04 datasets, which contain 300 and 448 recordings, respectively. Machine learning classifiers were tested using 5x5-fold cross-validation and feature selection. Results indicate that the proposed harmony-based rules combined with the timbral descriptor-based genre classification syst em lead to improved genre classification rates.