A Comparative Study on Polyphonic Musical Time Series Using MCMC Methods
A general harmonic model for pitch tracking of polyphonic musical time series will be introduced. Based on a model of Davy and Godsill (2002) the fundamental frequencies of polyphonic sound are estimated simultaneously. For an improvement of these results a preprocessing step was be implemented to build an extended polyphonic model.
All methods are applied on real audio data from the McGill University Master Samples (Opolko and Wapnick (1987)).
KeywordsFundamental Frequency Audio Signal MCMC Method MCMC Algorithm Correct Note
Unable to display preview. Download preview PDF.
- DAVY, M. and GODSILL, S. J. (2002): Bayesian Harmonic Models for Musical Pitch Estima-tion and Analysis. Technical Report 431, Cambridge University Engineering Department. GILKS, W. R., RICHARDSON, S. and SPIEGELHALTER D. J. (1996): Markov Chain Monte Carlo in Practice, Chapman & Hall.Google Scholar
- OPOLKO, F. and WAPNICK, J. (1987): McGill University Master Samples [Compact disc]: Montreal, Quebec: McGill University.Google Scholar