A sensor interface based on sparse NMF for piano musical transcription

A new sensor interface for detecting notes in the musical audio signals is proposed. Our sensor interface is able to transform acoustic pressure, caused by a sound wave of polyphonic piano music, into a musical score, that is a symbolic representation of musical notes. We focuses on note attack instant (onset) and note name (pitch). Signal processing techniques based on the Constant-Q Transform (CQT) are used to create a time-frequency representation of the signal. In particular, we propose a supervised classification methods based on sparse Non-Negative Matrix Factorization (NMF) to detect note pitch. We present a collection of experiments using the Inventions and Sinfonias BWV 772-801 of the composer J. S. Bach.