Cross-associating unlabelled timbre distributions to create expressive musical mappings

Dan Stowell, Mark D. Plumbley
Proceedings of the First Workshop on Applications of Pattern Analysis, PMLR 11:28-35, 2010.

Abstract

In timbre remapping applications such as concatenative synthesis, an audio signal is used as a template, and a mapping process derives control data for some audio synthesis algorithm such that it produces a new audio signal approximating the perceived trajectory of the original sound. Timbre is a multidimensional attribute with interactions between dimensions, and the control and synthesised signals typically represent sounds with different timbral ranges, so it is non-trivial to design a search process which makes best use of the timbral variety available in the synthesiser. We first discuss our preliminary work applying standard machine-learning techniques for this purpose (PCA, self-organising maps), and the reasons they were not satisfactory. We then describe a novel regression-tree technique which learns associations between unlabelled multidimensional timbre distributions.

Cite this Paper


BibTeX
@InProceedings{pmlr-v11-stowell10a, title = {Cross-associating unlabelled timbre distributions to create expressive musical mappings}, author = {Stowell, Dan and Plumbley, Mark D.}, booktitle = {Proceedings of the First Workshop on Applications of Pattern Analysis}, pages = {28--35}, year = {2010}, editor = {Diethe, Tom and Cristianini, Nello and Shawe-Taylor, John}, volume = {11}, series = {Proceedings of Machine Learning Research}, address = {Cumberland Lodge, Windsor, UK}, month = {01--03 Sep}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v11/stowell10a/stowell10a.pdf}, url = {https://proceedings.mlr.press/v11/stowell10a.html}, abstract = {In timbre remapping applications such as concatenative synthesis, an audio signal is used as a template, and a mapping process derives control data for some audio synthesis algorithm such that it produces a new audio signal approximating the perceived trajectory of the original sound. Timbre is a multidimensional attribute with interactions between dimensions, and the control and synthesised signals typically represent sounds with different timbral ranges, so it is non-trivial to design a search process which makes best use of the timbral variety available in the synthesiser. We first discuss our preliminary work applying standard machine-learning techniques for this purpose (PCA, self-organising maps), and the reasons they were not satisfactory. We then describe a novel regression-tree technique which learns associations between unlabelled multidimensional timbre distributions.} }
Endnote
%0 Conference Paper %T Cross-associating unlabelled timbre distributions to create expressive musical mappings %A Dan Stowell %A Mark D. Plumbley %B Proceedings of the First Workshop on Applications of Pattern Analysis %C Proceedings of Machine Learning Research %D 2010 %E Tom Diethe %E Nello Cristianini %E John Shawe-Taylor %F pmlr-v11-stowell10a %I PMLR %P 28--35 %U https://proceedings.mlr.press/v11/stowell10a.html %V 11 %X In timbre remapping applications such as concatenative synthesis, an audio signal is used as a template, and a mapping process derives control data for some audio synthesis algorithm such that it produces a new audio signal approximating the perceived trajectory of the original sound. Timbre is a multidimensional attribute with interactions between dimensions, and the control and synthesised signals typically represent sounds with different timbral ranges, so it is non-trivial to design a search process which makes best use of the timbral variety available in the synthesiser. We first discuss our preliminary work applying standard machine-learning techniques for this purpose (PCA, self-organising maps), and the reasons they were not satisfactory. We then describe a novel regression-tree technique which learns associations between unlabelled multidimensional timbre distributions.
RIS
TY - CPAPER TI - Cross-associating unlabelled timbre distributions to create expressive musical mappings AU - Dan Stowell AU - Mark D. Plumbley BT - Proceedings of the First Workshop on Applications of Pattern Analysis DA - 2010/09/30 ED - Tom Diethe ED - Nello Cristianini ED - John Shawe-Taylor ID - pmlr-v11-stowell10a PB - PMLR DP - Proceedings of Machine Learning Research VL - 11 SP - 28 EP - 35 L1 - http://proceedings.mlr.press/v11/stowell10a/stowell10a.pdf UR - https://proceedings.mlr.press/v11/stowell10a.html AB - In timbre remapping applications such as concatenative synthesis, an audio signal is used as a template, and a mapping process derives control data for some audio synthesis algorithm such that it produces a new audio signal approximating the perceived trajectory of the original sound. Timbre is a multidimensional attribute with interactions between dimensions, and the control and synthesised signals typically represent sounds with different timbral ranges, so it is non-trivial to design a search process which makes best use of the timbral variety available in the synthesiser. We first discuss our preliminary work applying standard machine-learning techniques for this purpose (PCA, self-organising maps), and the reasons they were not satisfactory. We then describe a novel regression-tree technique which learns associations between unlabelled multidimensional timbre distributions. ER -
APA
Stowell, D. & Plumbley, M.D.. (2010). Cross-associating unlabelled timbre distributions to create expressive musical mappings. Proceedings of the First Workshop on Applications of Pattern Analysis, in Proceedings of Machine Learning Research 11:28-35 Available from https://proceedings.mlr.press/v11/stowell10a.html.

Related Material