Monaural Musical Sound Separation

MMSS Audio Examples

This page demonstrates the monaural musical sound separation (MMSS) algorithm proposed by Y. Li, J. Woodruff and D.L. Wang as well as a comparison algorithm by Virtanen. For details of the proposed algorithm see:

Li Y., Woodruff J., and Wang D.L. (2009): Monaural musical sound separation based on pitch and common amplitude modulation. IEEE Transactions on Audio, Speech, and Language Processing, vol. 17, pp. 1361-1371.

Two Instruments:

Example One:
Original Sources: source 1; source 2
Mixture: mix
Proposed System: source 1; source 2
Virtanen 2006 System: source 1; source 2

Example Two:
Original Sources: source 1; source 2
Mixture: mix
Proposed System: source 1; source 2
Virtanen 2006 System: source 1; source 2

Three Instruments:

Original Sources: source 1; source 2; source 3
Mixture: mix
Estimated Sources: source 1; source 2; source 3