Paper Info
Title
Reconstructing completely overlapped notes from musical mixtures
Abstract
In mixtures of musical sounds, the problem of overlapped harmonics poses a significant challenge to source separation. Common Amplitude Modulation (CAM) is one of the most effective methods to resolve this problem. It, however, relies on non-overlapped harmonics from the same note being available. We propose an alternate technique for harmonic envelope estimation, based on Harmonic Temporal Envelope Similarity (HTES). We learn a harmonic envelope model for each instrument from the non-overlapped harmonics 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 better than an existing system based on CAM when the harmonics of pitched instruments are strongly overlapped.
Year
DOI
Venue
2011
10.1109/ICASSP.2011.5946387
ICASSP
Keywords
Field
DocType
musical sounds,overlapped note reconstruction,harmonic envelope estimation,common amplitude modulation,pitched instruments,music,note nonoverlapped harmonics,musical mixtures,harmonic temporal envelope similarity,overlapped harmonics,source separation,acoustic signal processing,music source separation,amplitude modulation,cam,signal reconstruction,musical instruments,htes,computer aided manufacturing,estimation,polynomials,harmonic analysis
Computer-aided manufacturing,Polynomial,Computer science,Harmonic analysis,Harmonics,Artificial intelligence,Source separation,Pattern recognition,Algorithm,Harmonic,Speech recognition,Amplitude modulation,Signal reconstruction
Conference
ISSN
ISBN
Citations 
1520-6149 E-ISBN : 978-1-4577-0537-3
978-1-4577-0537-3
3
PageRank 
References 
Authors
0.45
4
2
Name
Order
Citations
PageRank
Jinyu Han1697.92
Bryan Pardo283063.92