Identifying saxophonists from their playing styles

TitleIdentifying saxophonists from their playing styles
Publication TypeConference Paper
Year of Publication2007
Conference Name30th AES Conference
AuthorsRamirez, R., Maestre E., & Pertusa A.
AbstractThis paper describes a machine learning approach to the problem of identifying professional musicians from their playing style. We focus on the identification of jazz saxophonists by studying their expressive characteristics. In particular, we investigate expressive deviations of parameters such as pitch, timing, amplitude and timbre in monophonic audio recordings. We describe how we extract a symbolic description from the audio recordings and how we use this symbolic description to train a performance-based interpreter classifier.