Fast bayesian nmf algorithms enforcing harmonicity and temporal continuity in polyphonic music transcription

This article presents theoretical and experimental results about constrained non-negative matrix factorization (NMF) in a Bayesian framework, enforcing both spectral harmonicity and temporal continuity. We exhibit fast multiplicative update rules to perform the decomposition, which are then applied to perform polyphonic piano music transcription. This approach is shown to outperform other standard NMF-based transcription systems, providing a meaningful mid-level representation of the data.

[1]  Matija Marolt,et al.  A connectionist approach to automatic transcription of polyphonic piano music , 2004, IEEE Transactions on Multimedia.

[2]  P. Smaragdis,et al.  Non-negative matrix factorization for polyphonic music transcription , 2003, 2003 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (IEEE Cat. No.03TH8684).

[3]  Roland Badeau,et al.  Automatic transcription of piano music based on HMM tracking of jointly-estimated pitches , 2008, 2008 16th European Signal Processing Conference.

[4]  H. Sebastian Seung,et al.  Learning the parts of objects by non-negative matrix factorization , 1999, Nature.

[5]  Nancy Bertin,et al.  Nonnegative Matrix Factorization with the Itakura-Saito Divergence: With Application to Music Analysis , 2009, Neural Computation.

[6]  Emmanuel Vincent,et al.  Harmonic and inharmonic Nonnegative Matrix Factorization for Polyphonic Pitch transcription , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[7]  Tuomas Virtanen,et al.  Monaural Sound Source Separation by Nonnegative Matrix Factorization With Temporal Continuity and Sparseness Criteria , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[8]  Emmanuel Vincent,et al.  Enforcing Harmonicity and Smoothness in Bayesian Non-Negative Matrix Factorization Applied to Polyphonic Music Transcription , 2010, IEEE Transactions on Audio, Speech, and Language Processing.