Learning musical pitch structures with hierarchical hidden Markov models

Michèle Weiland (1) , Alan Smaill (1) , Peter Nelson (1)

1 University of Edinburgh

In this paper we attempt to demonstrate the strengths of Hierarchical Hidden Markov Models (HHMMs) in the representation and modelling of musical structures. We show how relatively simple HHMMs, containing a minimum of expert knowledge, use their advantage of having multiple layers to perform well on tasks where flat Hidden Markov Models (HMMs) struggle. The examples in this paper show a HHMM's performance at extracting higherlevel musical properties through the construction of simple pitch sequences, correctly representing the data set on which it was trained.

Domains: Computer Science [cs]; Musicology and performing arts

Michèle Weiland, Alan Smaill, Peter Nelson. Learning musical pitch structures with hierarchical hidden Markov models. Journées d'Informatique Musicale 2005, Association Française d'Informatique Musicale; Centre de recherche en Informatique et Création Musicale, Jun 2005, Saint-Denis, France. ⟨hal-03114882⟩