
Music Information Retrieval: Sound to Score and Music Theory
Michael Scott Asato Cuthbert, teaching MIT's 21M.383 Computational Music Theory and Analysis, introduces Music Information Retrieval as the problem of converting raw audio into musical scores. He lays out why the sound-to-score pipeline is not a simple transcription task but one tangled up with music theory itself, since decisions like how to notate rhythm, pitch, or harmonic structure require analytical judgment, not just signal processing. The lecture surveys the stages a system must pass through to turn an audio recording into readable notation and flags where ambiguity in music theory complicates automation. Aimed at students who already have some grounding in computational methods, it sets up MIR as a recurring theme for the rest of the course rather than a self-contained technical tutorial.