An automatic music transcription method is proposed. The method is based on a generative model that takes into account the translation of spectrum for an instrument and the sound path from the instrument to a microphone. The fundamental frequency (note), the spectrum of the instrument (basis pattern) and the sound path are estimated simultaneously using an extended complex nonnegative matrix factorization. The effectiveness of the proposed method is confirmed by synthetic data. © 2011 Springer-Verlag.
CITATION STYLE
Ikeuchi, R., & Ikeda, K. (2011). An automatic music transcription based on translation of spectrum and sound path estimation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7062 LNCS, pp. 532–540). https://doi.org/10.1007/978-3-642-24955-6_64
Mendeley helps you to discover research relevant for your work.