Does Inharmonicity Improve an NMF-Based Piano Transcription Model?

François Rigaud (1, 2) , Antoine Falaize (3) , B. David (1, 2) , Laurent Daudet (4)

1 S2A - Signal, Statistique et Apprentissage
2 TSI - Département Traitement du Signal et des Images
3 IRCAM - Institut de Recherche et Coordination Acoustique/Musique
4 ESPCI Paris - Ecole Superieure de Physique et de Chimie Industrielles de la Ville de Paris

This paper investigates how precise a model should be for a robust model-based NMF analysis of piano recordings. While inharmonicity is an essential feature of piano tones from a perceptual point of view, its explicit inclusion in sound models is not straightforward and may even damage the quality of the analysis. Here, we assess the quality of the analysis with a transcription task, and compare three different models for the spectra of the dictionary : one strictly harmonic, one following the theoretical inharmonicity law, and one with relaxed inharmonicity constraints. Experimental results show that both inharmonic models can indeed significantly enhance the results, but only in the case when a good initialization is provided.

Keywords: non-negative matrix factorization; music transcription; piano; inharmonicity

Domains: Signal and Image processing; Acoustics [physics.class-ph]

François Rigaud, Antoine Falaize, B. David, Laurent Daudet. Does Inharmonicity Improve an NMF-Based Piano Transcription Model?. 38th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2013), May 2013, Vancouver, Canada. pp.11-15. ⟨hal-00856734⟩