Context-Aware Hidden Markov Models of Jazz Music with Variable Markov Oracle

Cheng-I Wang, Shlomo Dubnov
2017 Zenodo  
In this paper, a latent variable model based on the Variable Markov Oracle (VMO) is proposed to capture long-term temporal relationships between sequential observations. The proposed latent variable model is a multi-level Hidden Markov Model (HMM) extracted from the VMO, and is called the VMO-HMM. The musical importance of the proposed model is demonstrated by modeling harmonic progressions of jazz music. The VMO-HMM is able to reveal functional harmony relations beyond chord labels, provides
more » ... eoretic insight into scale-chord relations and aids jazz harmony improvisation as a model of music meta-creation.
doi:10.5281/zenodo.4285246 fatcat:cgpfd52jpzgebnbx2v2rirylzm