Polyphonic music transcription through dynamic networks and spectral pattern identification ∗

The automatic extraction of the notes that were played in a digital musical signal (automatic music transcription) is an open problem. A number of techniques have been applied to solve it without concluding results. This work tries to pose it through the identification of the spectral pattern of a given instrument in the signal spectrogram using time-delay neural networks. We will work in the monotimbrical polyphonic version of the problem: more than one note can sound at the same time but always played by just one instrument. Our purpose is to discover wether a neural network fed only with an spectrogram can detect the notes of a polyphonic music score. In this paper our preliminary but promising results using synthetic instruments are presented.

[1]  James Anderson Moorer,et al.  On the segmentation and analysis of continuous musical sound by digital computer , 1975 .

[2]  W. Hess.  Algorithms and devices for pitch determination of speech signals. , 1982, Phonetica.

[3]  Geoffrey E. Hinton,et al.  Learning representations by back-propagating errors , 1986, Nature.

[4]  Editors , 1986, Brain Research Bulletin.

[5]  Geoffrey E. Hinton,et al.  A time-delay neural network architecture for isolated word recognition , 1990, Neural Networks.

[6]  Anders Krogh,et al.  Introduction to the theory of neural computation , 1994, The advanced book program.

[7]  Judith C. Brown,et al.  Musical frequency tracking using the methods of conventional and , 1991 .

[8]  Louis P. DiPalma,et al.  Music and Connectionism , 1991 .

[9]  Roland Wilson,et al.  Note recognition in polyphonic music using neural networks , 1993 .

[10]  D.R. Hush,et al.  Progress in supervised neural networks , 1993, IEEE Signal Processing Magazine.

[11]  Dik J. Hermes,et al.  Pitch analysis , 1993 .

[12]  Keith D. Martin,et al.  A Blackboard System for Automatic Transcription of Simple Polyphonic Music , 1996 .

[13]  Xavier Rodet,et al.  Timbre Recognition with Combined Stationary and Temporal Features , 1998, ICMC.

[14]  Ichiro Fujinaga,et al.  Machine recognition of timbre using steady-state tone of acoustic musical instruments , 1998, ICMC.

[15]  Matija Marolt.  Transcription of polyphonic piano music with neural networks , 2000, 2000 10th Mediterranean Electrotechnical Conference. Information Technology and Electrotechnology for the Mediterranean Countries. Proceedings. MeleCon 2000 (Cat. No.00CH37099).

[16]  Anssi Klapuri,et al.  AUTOMATIC TRANSCRIPTION OF MUSIC , 2003 .