Modelling Symbolic Music: Beyond the Piano Roll

Christian Walder
Proceedings of The 8th Asian Conference on Machine Learning, PMLR 63:174-189, 2016.

Abstract

In this paper, we consider the problem of probabilistically modelling symbolic music data. We introduce a representation which reduces polyphonic music to a univariate categorical sequence. In this way, we are able to apply state of the art natural language processing techniques, namely the long short-term memory sequence model. The representation we employ permits arbitrary rhythmic structure, which we assume to be given. We show that our model is effective on all four piano roll based benchmark datasets. We further improve our model by augmenting our training data set with transpositions of the original pieces through all musical keys, thereby convincingly advancing the state of the art on these benchmark problems. We also fit models to music which is unconstrained in its rhythmic structure, discuss the properties of this model, and provide musical samples which are more sophisticated than previously possible with this class of recurrent neural network sequence models. We also provide our newly preprocessed data set of non piano-roll music data. To facilitate future work we describe and provide a new carefully preprocessed dataset of 19700 classical midi music files — significantly more than previously available.

Cite this Paper

BibTeX
@InProceedings{pmlr-v63-walder88, title = {Modelling Symbolic Music: Beyond the Piano Roll}, author = {Walder, Christian}, booktitle = {Proceedings of The 8th Asian Conference on Machine Learning}, pages = {174--189}, year = {2016}, editor = {Durrant, Robert J. and Kim, Kee-Eung}, volume = {63}, series = {Proceedings of Machine Learning Research}, address = {The University of Waikato, Hamilton, New Zealand}, month = {16--18 Nov}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v63/walder88.pdf}, url = {https://proceedings.mlr.press/v63/walder88.html}, abstract = {In this paper, we consider the problem of probabilistically modelling symbolic music data. We introduce a representation which reduces polyphonic music to a univariate categorical sequence. In this way, we are able to apply state of the art natural language processing techniques, namely the long short-term memory sequence model. The representation we employ permits arbitrary rhythmic structure, which we assume to be given. We show that our model is effective on all four piano roll based benchmark datasets. We further improve our model by augmenting our training data set with transpositions of the original pieces through all musical keys, thereby convincingly advancing the state of the art on these benchmark problems. We also fit models to music which is unconstrained in its rhythmic structure, discuss the properties of this model, and provide musical samples which are more sophisticated than previously possible with this class of recurrent neural network sequence models. We also provide our newly preprocessed data set of non piano-roll music data. To facilitate future work we describe and provide a new carefully preprocessed dataset of 19700 classical midi music files — significantly more than previously available.} }
Endnote
%0 Conference Paper %T Modelling Symbolic Music: Beyond the Piano Roll %A Christian Walder %B Proceedings of The 8th Asian Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2016 %E Robert J. Durrant %E Kee-Eung Kim %F pmlr-v63-walder88 %I PMLR %P 174--189 %U https://proceedings.mlr.press/v63/walder88.html %V 63 %X In this paper, we consider the problem of probabilistically modelling symbolic music data. We introduce a representation which reduces polyphonic music to a univariate categorical sequence. In this way, we are able to apply state of the art natural language processing techniques, namely the long short-term memory sequence model. The representation we employ permits arbitrary rhythmic structure, which we assume to be given. We show that our model is effective on all four piano roll based benchmark datasets. We further improve our model by augmenting our training data set with transpositions of the original pieces through all musical keys, thereby convincingly advancing the state of the art on these benchmark problems. We also fit models to music which is unconstrained in its rhythmic structure, discuss the properties of this model, and provide musical samples which are more sophisticated than previously possible with this class of recurrent neural network sequence models. We also provide our newly preprocessed data set of non piano-roll music data. To facilitate future work we describe and provide a new carefully preprocessed dataset of 19700 classical midi music files — significantly more than previously available.
RIS
TY - CPAPER TI - Modelling Symbolic Music: Beyond the Piano Roll AU - Christian Walder BT - Proceedings of The 8th Asian Conference on Machine Learning DA - 2016/11/20 ED - Robert J. Durrant ED - Kee-Eung Kim ID - pmlr-v63-walder88 PB - PMLR DP - Proceedings of Machine Learning Research VL - 63 SP - 174 EP - 189 L1 - http://proceedings.mlr.press/v63/walder88.pdf UR - https://proceedings.mlr.press/v63/walder88.html AB - In this paper, we consider the problem of probabilistically modelling symbolic music data. We introduce a representation which reduces polyphonic music to a univariate categorical sequence. In this way, we are able to apply state of the art natural language processing techniques, namely the long short-term memory sequence model. The representation we employ permits arbitrary rhythmic structure, which we assume to be given. We show that our model is effective on all four piano roll based benchmark datasets. We further improve our model by augmenting our training data set with transpositions of the original pieces through all musical keys, thereby convincingly advancing the state of the art on these benchmark problems. We also fit models to music which is unconstrained in its rhythmic structure, discuss the properties of this model, and provide musical samples which are more sophisticated than previously possible with this class of recurrent neural network sequence models. We also provide our newly preprocessed data set of non piano-roll music data. To facilitate future work we describe and provide a new carefully preprocessed dataset of 19700 classical midi music files — significantly more than previously available. ER -
APA
Walder, C.. (2016). Modelling Symbolic Music: Beyond the Piano Roll. Proceedings of The 8th Asian Conference on Machine Learning, in Proceedings of Machine Learning Research 63:174-189 Available from https://proceedings.mlr.press/v63/walder88.html.

Related Material