Music Information Retrieval using Scikit-learn

Music information retrieval (MIR) is an interdisciplinary field bridging the domains of statistics, signal processing, machine learning, musicology, biology, and more.  MIR algorithms allow a computer to make sense of audio data in order to bridge the semantic gap between high-level musical information — e.g. tempo, key, pitch, instrumentation, chord progression, genre, song structure — and low-level audio data.

In this talk, Steve Tjoa from Humtap surveys common research problems in MIR, including music fingerprinting, transcription, classification, and recommendation, and recently proposed solutions in the research literature. Steve gives a high-level overview as well as concrete examples (and a live demo) of implementing and evaluating MIR algorithms in Python using Scikit-learn and the IPython notebook.


This talk was given at the SF Bay Area Machine Learning meetup at Thumbtack.