Transientstudent

Papers

Bello, Juan Pablo, Guiliano Monti, and Mark Sandler. 2000. "Techniques for Automatic Music Transcription." In Proceedings of the First International Conference on Music Information Retrieval (ISMIR), Plymouth, Massachusetts. Available online at http://ciir.cs.umass.edu/music2000/papers/bello_paper.pdf (accessed 1 March, 2007).
Evaluation of two systems for Music Transcription: Monophonic autocorrelation and Polyphonic transcription using a blackboard system approach. The Blackboard system uses a neural network chord recognizer that can dynamically reconfigure the system as it 'learns.' Builds on the system proposed by Martin (1996) but the addition of a neural network component yields more accurate results on a greater diversity of input.

Bello, Juan Pablo, and Mark Sandler. 2000. "Blackboard System and Top-Down Processing for the Transcription of Simple Polyphonic Music." In Proceedings of the COST G-6 Conference on Digital Audio Effects (DAFX-00), Verona, Italy. Available online at http://profs.sci.univr.it/~dafx/Final-Papers/pdf/Bello_DAFx2000.pdf (accessed 1 March, 2007).
Essentially the same system as the previous paper is described, but it is not compared to a monophonic autocorrelation system. As well, this paper is slightly less technical and offers diagrams and charts of actual system performance when presented with a polyphonic piano input.

Dixon, Simon. 2000. "On the Computer Recognition of Solo Piano Music." In Proceedings of the Australasian Computer Music Association Conference, Brisbane, Australia. Available online at CiteSeer (accessed 1 March, 2007).
A brief paper outlining a basic Piano Transcription system. Useful for a overview of the general process, but light on exact details on how their system is constructed or specific results.

Klapuri, Anssi, Tuomas Virtanen, Antti Eronen, and Jarno Seppanen. 2001. "Automatic Transcription of Musical Recordings." In Proceedings of the Consistent & Reliable Cues Workshop (CRAC-01), Aalborg, Denmark. Available online at http://www.cs.tut.fi/sgn/arg/klap/crac2001/crac2001.pdf (accessed 1 March, 2007).
Not piano specific, but presents some of the challenges of polyphonic music transcription such as the elimination of non-harmonic (i.e. some percussion) instruments. Also mentions the problems with detecting percussion instruments (much of their energy is translated into some form of harmonic energy) and incorporating musical instrument modelling into their system to improve recognition.

Marolt, Matija. 2005. "A Connectionist Model of Finding Partial Groups in Music Recordings With Application to Music Transcription." In Proceedings of the Seventh International Conference on Adaptive and Natural Computing Algorithms, Coimbra, Portuagal. Available online at http://lgm.fri.uni-lj.si/~matic/clanki/icannga2005_marolt.pdf (accessed 1 March, 2007).
A system for polyphonic piano transcription is presented, using a neural network approach. In this network, adaptive oscillators are used to recognize partials in given pitches. This information is then used to build a fundamental pitch on the idea that a given set of partials is unique to a single fundamental pitch. By recognizing all the partials present in an audio stream a chord can be recognized.

Marolt, Matija. 2004. "A Connectionist Approach to Automatic Transcription of Polyphonic Piano Music." IEEE Transactions on Multimedia 6 (3): 494-95. Available online at http://lgm.fri.uni-lj.si/~matic/clanki/ieee.tmm.transcription.pdf (accessed 1 March, 2007). Describes the previous neural network implementation in greater detail. Gives further detail on the partial tracking mechanism, the 'auditory model' used by the system to 'hear' sounds, and a fairly comprehensive section on experiment results.

Martin, Keith D. 1996. A Blackboard System for Automatic Transcription of Simple Polyphonic Music. Mit Media Laboratory Perceptual Computing Section Technical Report No. 385. Available online at http://alumni.media.mit.edu/~kdm/research/papers/kdm-TR385.pdf (accessed 1 March, 2007).
One of the first, if not the first, uses of a blackboard model for polyphonic music transcription. Uses a static group of Knowledge Sources to analyse information on the 'blackboard' and arrive at a note and chord hypothesis. This paper goes into some detail about the specific roles of each of the knowledge sources, and is a good overview of blackboard systems in this context.

Monti, Guiliano, and Mark Sandler. 2002. "Automatic Polyphonic Piano Note Extraction Using Fuzzy Logic in a Blackboard System." In Proceedings of the 5th International Conference on Digital Audio Effects (DAFx-02), Hamburg, Germany. Available online at The author's website (accessed 1 March, 2007).
A blackboard system using a Fuzzy Inference System (FIS) as one of the knowledge sources. This FIS calculates a fundamental note's probability of being located in a chord based on an analysis of the spectral peaks present in the audio analysis.

Moorer, James A. 1977. "On transcription of musical sound by computer." Computer Music Journal 1(4):32-38.
The first practical method of polyphonic music transcription. Severly restricted in what input it accepts, but an important first step.

Poliner, Graham E., and Daniel P. W. Ellis. 2007. "A Discriminative Model for Polyphonic Piano Transcription." EURASIP Journal on Advances in Signal Processing v. 2007. Available online at http://www.ee.columbia.edu/~dpwe/pubs/PoliE06-piano.pdf (accessed 1 March, 2007).
Outlines an approach to piano transcription based on Support Vector Machines and Hidden Markov Models.

Poliner, Graham E., and Daniel P. W. Ellis. 2005. "A Classification Approach to Melody Transcription." In Proceedings of the Sixth Annual Conference on Music Information Retrieval (ISMIR), London, UK. Available online at http://www.ee.columbia.edu/~dpwe/pubs/ismir05-melody.pdf (accessed 1 March, 2007).
Further details on the SVM and HMM model system of transcription is given. Concludes that a system based entirely on machine learning with a training set can successfully 'learn' musical structure with no in-built model of musical knowledge.

Raphael, Christopher. 2002. "Automatic Transcription of Piano Music." In Proceedings of the Third International Conference on Music Information Retrieval (ISMIR), Paris, France. Availble online at http://xavier.informatics.indiana.edu/~craphael/papers/ismir02_rev.pdf (accessed 1 March, 2007).
Another HMM approach to piano transcription. This model uses a signal-score method to form hypotheses for notes, and can be trained in an unsupervised fashion.

Websites & Online Resources

Piano Transcription presentation by Catherine Lai

A Brief History of Connectionism.

Blackboard systems (includes links to distributed reasoning, etc.)