TY - GEN
T1 - Multiple-f0 tracking based on a high-order hmm model
AU - Chang, Wei Chen
AU - Su, Alvin W.Y.
AU - Yeh, Chunghsin
AU - Roebel, Axel
AU - Rodet, Xavier
N1 - Funding Information:
The author would like to thank the Cosmic Refrigeration Private Ltd, Pune, for providing the test facility. Special thanks to Mr. Shivaprasad Kalluraya and Mr. Sharan Challamarad for their kind cooperation and assistance during experimental work. We would like to acknowledge ASHRAE, Atlanta, GA for their financial support under student project grant for building prototype and conduction of the experimental work.
PY - 2008
Y1 - 2008
N2 - This paper is about multiple-F0 tracking and the estimation of the number of harmonic source streams in music sound signals. A source stream is understood as generated from a note played by a musical instrument. A note is described by a hidden Markov model (HMM) having two states: the attack state and the sustain state. It is proposed to first perform the tracking of F0 candidates using a high-order hidden Markov model, based on a forward-backward dynamic programming scheme. The propagated weights are calculated in the forward tracking stage, followed by an iterative tracking of the most likely trajectories in the backward tracking stage. Then, the estimation of the underlying source streams is carried out by means of iteratively pruning the candidate trajectories in a maximum likelihood manner. The proposed system is evaluated by a specially constructed polyphonic music database. Compared with the frame-based estimation systems, the tracking mechanism improves significantly the accuracy rate.
AB - This paper is about multiple-F0 tracking and the estimation of the number of harmonic source streams in music sound signals. A source stream is understood as generated from a note played by a musical instrument. A note is described by a hidden Markov model (HMM) having two states: the attack state and the sustain state. It is proposed to first perform the tracking of F0 candidates using a high-order hidden Markov model, based on a forward-backward dynamic programming scheme. The propagated weights are calculated in the forward tracking stage, followed by an iterative tracking of the most likely trajectories in the backward tracking stage. Then, the estimation of the underlying source streams is carried out by means of iteratively pruning the candidate trajectories in a maximum likelihood manner. The proposed system is evaluated by a specially constructed polyphonic music database. Compared with the frame-based estimation systems, the tracking mechanism improves significantly the accuracy rate.
UR - http://www.scopus.com/inward/record.url?scp=77955816771&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=77955816771&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:77955816771
SN - 9789512295173
T3 - Proceedings of the International Conference on Digital Audio Effects, DAFx
SP - 379
EP - 386
BT - Proceedings - 11th International Conference on Digital Audio Effects, DAFx 2008
T2 - 11th International Conference on Digital Audio Effects, DAFx 2008
Y2 - 1 September 2008 through 4 September 2008
ER -