Synthesis of Sound Textures by Learning and Resampling of Wavelet Trees                                     

Shlomo Dubnov                                         Ziv Bar-Joseph
CommunicationSystems Engineering                       Laboratoryfor ComputerScience
Ben-GurionUniversity, Beer-Sheva, Israel                       MIT, Cambridge,MA
Ran El-Yaniv                         Dani Lischinski            Michael Werman
Departmentof ComputerScience                 School of ComputerScience andEngineering
Technion Israel Instituteof Technology             The Hebrew University of Jerusalem,Israel



Abstract                             
In this paper we presen ta statistical learning algorithm for synthesizing
new random instances of a sound texture given an example of such a texture       
as input. A large class of natural and artificial sounds such as rain, waterfall, 
traffic noises, people babble, machine noises, etc., can be regarded as sound     
texturessound signals that are approximately stationary at some scale. 
Treating the input sound texture as a sample of a stochastic process, we  
construct a tree representing a hierarchical wavelet transform of the signal. 
From this tree, new randomtrees are generatedby learningand sampling      
the conditional probabilities of the paths in the original tree. Transformation 
of these random trees back into signals results in new sound textures  
that closely resemble the sonic impression of the original sound source but   
without exactly repeating it. Applications of this method are abundant and     
include, for example, automatic generation of sound effects, creative musical 
and sonic manipulations, and virtual reality sonification. Examples are  
visually demonstrated in the paper and acoustically demonstrated in an accompanying 
website.                                                       


References                                                                                 
[1]  Ziv Bar-Joseph, Ran El-Yaniv, Dani Lischinski, and Michael Werman. Texture 
mixing and texture movie synthesis using statistical learning. IEEE              
Transactions on Visualization and Computer Graphics, 7(2):120135,2001.                    
[2]  M. Basseville, A. Benveniste, K.C. Chou, S.A. Golden, R. Nikoukhah, and                  
A.S. Willsky. Modeling and estimation of multire solution stochastic processes. 
IEEE Transactions on Information Theory, 38(2):766784,1992.                     
[3]  M. Basseville, A. Benveniste, and A.S. Willsky. Multiscale autoregressive             
processes, part II: Lattice structures for whitening and modeling. IEEE
Transactions on Signal Processing, 40(8):19351954,1992.                                 
[4]  G.E.P. Box, G.M. Jenkins, G.C. Reinsel, andG. Jenkins. Time Series Analysis: 
Forecasting and Control. Prentice Hall, 1994.                                      
[5]  Ingrid Daubechies. Orhtonormal bases of compactly supported wavelets.                   
Communications on Pure and Applied Mathematics, 41(7):909996, October 1988.                                                                             
[6]  G. De Poli andA. Piccialli. Pitch-synchhronous granular synthesis. In Representation
of Musical Signals. The MIT Press, 1991.                                   
[7]  Ronald A. DeVore, Bjorn Jawerth, and Bradley J. Lucier. Image compression 
through wavelet transform coding. IEEE Transactions on Information Theory, 
38(2 (Part II)):719746, 1992.                                                
[8]  S. Dubnov, R. El-Yaniv, and G. Assayag. Universal classification applied to               
musical sequences. In Proceedings of International Computer Music Conference, 
New Jersey, 1998.                                                            
[9]  R.O. Duda and P.E. Hart. Pattern Classification and Scene Analysis. John                   
Wiley & Sons, 1973.                                                                   
[10]  Ran El-Yaniv, Shai Fine, and Naftali Tishby. Agnostic classification of 
Markovian sequences. In Advances in Neural Information Processing Sysems, 
volume 10. The MIT Press, 1998.                                                  
[11]  D. Gabor. Acoustical Quanta and the Theory of Hearing. In Nature, Vol.             
159, No. 4044, 1947                                                                   
[12]  N. Merhav and M. Feder. Universal prediction. IEEE Transactions on Information
Theory, 44(6):21242147,1998.                                                         
[13]  Curtis Roads. Introduction to granular synthesis. Computer Music Journal,                       
12(2):1113,1988.                                                                           
[14]  V.N. Vapnik. Statistical Learning Theory. John Wiley & Sons, 1998.                           
[15]  G.W. Wornell and A.V. Oppenheim. Wavelet-based representations for a                          
class of self-similar signals with application to fractal modulation. IEEE                   
Transactions on Information Theory, 38(2):785800,1992.                                        