Annotated Bibliography for the Class Presentation
on Automated Transcription of Polyphonic Piano Music
Dixon, S. 2000. On the Computer Recognition of Solo Piano Music. Australasian Computer Music Conference. 31-7.
Available: http://www.mikropol.net/volume6/dixon_s/mikro.html (Last accessed February 17, 2005)
This paper is one of the easier reading papers on the
work towards a computer system for the automatic transcription of piano performances.
Stanadard signal processing techniques based on the short time Fourier transform
are used to create time-frequency representation and classi peak findg algorithms
are described.
Marolt, M. 2004. A connectionist approach to automatic transcription
of polyphonic piano music. IEEE Transactions on Multimedia 6, no. 3 (June):
439-49.
Available: lgm.fri.uni-lj.si/~matic/
clanki/ieee.tmm.transcription.pdf (Last accessed February 17, 2005)
Marolt published many papers on the topic of automated
transcription system for piano music, but this is the best written in terms
of the technical detail and organization of the paper. This time he presented
a connectionist approach to automatic transcription of polyphonic piano music
using several neural networks where a new model for tracking partials in polyphonic
audio signal based on networks of adaptive oscillators was proposed.
Martin, K. 1996. A blackboard system for automatic transcription
of simple polyphonic music. MIT Media Laboratory Perceptual Computing Section
Technical Report No. 385.
This paper is cited in many of the related works areas
in papers. Martin's piano transcription allowed up to four voices in the input
data; however, it was restricted to the chorale style of J.S. Bach, where
notes have relatively long duration and change simultaneously .
Monti, G, and M. Sandler. 2002. Automatic Polyphonic Piano Note Extraction Using Fuzzy Logic in a Blackboard System. Proceedings of the International Conference on Digital Audio Effects. 39-44.
Monti and Sandler presented the implementation of Polyphonic
Note Recognition using a Fuzzy Inference System (FIS) as part of many of the
processes called Knowledge Sources (KSs) in a Blackboard system. The use of
this technique is considered to be one of the most robust among all the various
approaches taken by different research groups. I find the paper very hardest
to read, especially when it ends the with a result section without any summary
of the overall system or any concluding remarks.
Moorer, J. 1975. On the segmentation and analysis of continuous musical sound by digital computer. Ph.D. thesis, Stanford University, CCRMA.
This is always cited to be the pioneering work in the
area of polyphonic transcription, and the fact that the system has limitations
on note range and only supports two different timbre voices is always mentioned.
Raphael, C. 2002. Automatic Transcription of Piano Music.
Proceedings of the International Conference on Music Information Retrieval.
This paper uses the HMM approach.It seems difficult at
first, but after several times of reading the material is not all incomprehensible,
as in the case of Monti's.