It's Time for a Song - Transcribing Recordings of Bell-playing Clocks

The paper presents an algorithm for automatic transcription of recordings of bell-playing clocks. Bell-playing clocks are clocks containing a hidden bell-playing mechanism that is periodically activated to play a melody. Clocks from the eighteenth century give us unique insight into the musical taste of their owners, so we are interested in studying their repertoire and performances - thus the need for automatic transcription. In the paper, we first present an analysis of acoustical properties of bells found in bell-playing clocks. We propose a model that describes positions of bell partials and an algorithm that discovers the number of bells and positions of their partials in a given recording. To transcribe a recording, we developed a probabilistic method that maximizes the joint probability of a note sequence given the recording and positions of bell partials. Finally, we evaluate our algorithms on a set of recordings of bell-playing clocks.

[1]  Mark D. Plumbley,et al.  Unsupervised analysis of polyphonic music by sparse coding , 2006, IEEE Transactions on Neural Networks.

[2]  Judith C. Brown.  Calculation of a constant Q spectral transform , 1991 .

[3]  Emmanuel Vincent,et al.  Low Bit-Rate Object Coding of Musical Audio Using Bayesian Harmonic Models , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[4]  P. Smaragdis,et al.  Non-negative matrix factorization for polyphonic music transcription , 2003, 2003 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (IEEE Cat. No.03TH8684).

[5]  Hyunsoo Kim,et al.  Nonnegative Matrix Factorization Based on Alternating Nonnegativity Constrained Least Squares and Active Set Method , 2008, SIAM J. Matrix Anal. Appl..

[6]  D. Temperley.  Music and probability , 2006 .

[7]  M. Marolt.  NON-NEGATIVE MATRIX FACTORIZATION WITH SELECTIVE SPARSITY CONSTRAINTS FOR TRANSCRIPTION OF BELL CHIMING RECORDINGS , 2009 .

[8]  Anssi Klapuri,et al.  Signal Processing Methods for Music Transcription , 2006 .

[9]  William A. Hibbert,et al.  The Quantification of Strike Pitch and Pitch Shifts in Church Bells , 2008 .

[10]  Shigeki Sagayama,et al.  Multipitch Analysis with Harmonic Nonnegative Matrix Approximation , 2007, ISMIR.

[11]  P. C. Pandey,et al.  The Journal of the Acoustical Society of America , 1939 .

[12]  Tuomas Virtanen,et al.  Monaural Sound Source Separation by Nonnegative Matrix Factorization With Temporal Continuity and Sparseness Criteria , 2007, IEEE Transactions on Audio, Speech, and Language Processing.

[13]  Bernhard Niedermayer.  Non-Negative Matrix Division for the Automatic Transcription of Polyphonic Music , 2008, ISMIR.

[14]  Emmanuel Vincent,et al.  Harmonic and inharmonic Nonnegative Matrix Factorization for Polyphonic Pitch transcription , 2008, 2008 IEEE International Conference on Acoustics, Speech and Signal Processing.

[15]  Mark B. Sandler,et al.  On the use of phase and energy for musical onset detection in the complex domain , 2004, IEEE Signal Processing Letters.