Automatic Transcription of Piano Music by Sparse Representation of Magnitude Spectra

C.-T. Lee, Y.-H. Yang, and Homer H. Chen, "Automatic Transcription of Piano Music by Sparse Representation of Magnitude Spectra,"
IEEE Int. Conf. Multimedia and Expo. 2011 (ICME'11), accepted.

[full text available after the conference]

Music transcription is the process of converting a musical recording to a musical score. It is useful for content-based music retrieval, such as query by example, music similarity measurement, and cover song identification. This project mainly deals with piano music transcription.

Assuming that the waveforms of piano notes are pre-stored and that the magnitude spectrum of a piano signal segment can be represented as a linear combination of the magnitude spectra of the pre-stored piano waveforms, we formulate the automatic transcription of polyphonic piano music as a sparse representation problem. First, the note candidates of the piano signal segment are found by using heuristic rules. Then, the sparse representation problem is solved by l1-regularized minimization, followed by temporal smoothing the frame-level results based on hidden Markov models. Evaluation against three state-of-the-art systems using ten classical music recordings of a real piano is performed to show the performance improvement of the proposed system.

Data Sets

The piano music recordings are downloaded from LabROSA's project page. Ten classical music recordings are used.


Here we provide the transcribed version in MIDI format. As for the original WAV and corresponding MIDI (ground truth) files, please visit LabROSA's page.
BachPrelude and Fugue No.2 in C MinorTranscription
BeethovenSonata no. 8 Pathetique in C minor, 3rd movementTranscription
MozartSonata K.333 in Bb Major, 1st MovementTranscription
SchumannScenes from Childhood No.4Transcription

Any feedbacks or comments are more than welcomed!