RECONSTRUCTING COMPLETELY OVERLAPPED NOTES FROM MUSICAL MIXTURES

In mixtures of musical sounds, the problem of overlapped har-monics poses a significant challenge to source separation. Common Amplitude Modulation (CAM) is one of the most effective meth-ods to resolve this problem. It, however, relies on non-overlapped harmonics from the same note being available. We propose an al-ternate technique for harmonic envelope estimation, based on Har-monic Temporal Envelope Similarity (HTES). We learn a harmonic envelope model for each instrument from the non-overlapped har-monics of notes of the same instrument, wherever they occur in the recording. This model is used to reconstruct the harmonic envelopes for overlapped harmonics. This allows reconstruction of completely overlapped notes. Experiments show our algorithm performs bet-ter than an existing system based on CAM when the harmonics of pitched instruments are strongly overlapped