Mirex 2009 a Multi-feature-set Multi-classifier Ensemble Approach for Audio Music Classification

The approach of combining a multitude of audio features and also symbolic features (through transcription of audio to MIDI) for music classification proved useful, as shown previously. We extended the system submitted to MIREX 2008 by including temporal audio features, adding another audio analysis algorithm based on finding templates on music, enhancing the polyphonic audio to MIDI transcription system and using an ensemble of classification models specializing on feature subsets, rather than combining all features to feed a single classifier, like in the previous MIREX. Recent research in music genre classification hints at a glass ceiling being reached using timbral audio features.