Generating Polyphonic Music Using Tied Parallel Networks

Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 10198)

Abstract

We describe a neural network architecture which enables prediction and composition of polyphonic music in a manner that preserves translation-invariance of the dataset. Specifically, we demonstrate training a probabilistic model of polyphonic music using a set of parallel, tied-weight recurrent networks, inspired by the structure of convolutional neural networks. This model is designed to be invariant to transpositions, but otherwise is intentionally given minimal information about the musical domain, and tasked with discovering patterns present in the source dataset. We present two versions of the model, denoted TP-LSTM-NADE and BALSTM, and also give methods for training the network and for generating novel music. This approach attains high performance at a musical prediction task and successfully creates note sequences which possess measure-level musical structure.

Notes

Acknowledgments

We would like to thank Dr. Robert Keller for helpful discussions and advice. We would also like to thank the developers of the Theano framework [20], which we used to run our experiments, as well as Harvey Mudd College for providing computing resources. This work used the Extreme Science and Engineering Discovery Environment (XSEDE) [23], which is supported by National Science Foundation grant number ACI-1053575.

References

  1. 1.
    Bellgard, M.I., Tsang, C.P.: Harmonizing music the boltzmann way. Connect. Sci. 6(2–3), 281–297 (1994)CrossRefGoogle Scholar
  2. 2.
    Bengio, S., Vinyals, O., Jaitly, N., Shazeer, N.: Scheduled sampling for sequence prediction with recurrent neural networks. In: Advances in Neural Information Processing Systems, pp. 1171–1179 (2015)Google Scholar
  3. 3.
    Boulanger-Lewandowski, N., Bengio, Y., Vincent, P.: Modeling temporal dependencies in high-dimensional sequences: application to polyphonic music generation and transcription. In: Proceedings of the 29th International Conference on Machine Learning (ICML-2012), pp. 1159–1166 (2012)Google Scholar
  4. 4.
    Eck, D., Schmidhuber, J.: A first look at music composition using LSTM recurrent neural networks. Istituto Dalle Molle Di Studi Sull Intelligenza Artificiale (2002)Google Scholar
  5. 5.
    Fernández, J.D., Vico, F.: AI methods in algorithmic composition: a comprehensive survey. J. Artif. Intell. Res. 48, 513–582 (2013)MathSciNetMATHGoogle Scholar
  6. 6.
    Greff, K., Srivastava, R.K., Koutnk, J., Steunebrink, B.R., Schmidhuber, J.: LSTM: A search space odyssey. arXiv preprint arXiv:1503.04069 (2015)
  7. 7.
    Hild, H., Feulner, J., Menzel, W.: HARMONET: a neural net for harmonizing chorales in the style of JS Bach. In: NIPS, pp. 267–274 (1991)Google Scholar
  8. 8.
    Hochreiter, S., Schmidhuber, J.: Long short-term memory. Neural comput. 9(8), 1735–1780 (1997)CrossRefGoogle Scholar
  9. 9.
    Kaiser, Ł., Sutskever, I.: Neural GPUs learn algorithms. arXiv preprint arXiv:1511.08228 (2015)
  10. 10.
    Kalchbrenner, N., Danihelka, I., Graves, A.: Grid long short-term memory. arXiv preprint arXiv:1507.01526 (2015)
  11. 11.
    Kingma, D.P., Welling, M.: Auto-encoding variational Bayes. arXiv preprint arXiv:1312.6114 (2013)
  12. 12.
    Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems, pp. 1097–1105 (2012)Google Scholar
  13. 13.
    Larochelle, H., Murray, I.: The neural autoregressive distribution estimator. In: International Conference on Artificial Intelligence and Statistics, pp. 29–37 (2011)Google Scholar
  14. 14.
    Lewis, J.P.: Creation by refinement and the problem of algorithmic music composition. In: Music and Connectionism, p. 212 (1991)Google Scholar
  15. 15.
    Moon, T., Choi, H., Lee, H., Song, I.: RnnDrop: a novel dropout for RNNs in ASR. In: Automatic Speech Recognition and Understanding (ASRU) (2015)Google Scholar
  16. 16.
    Mozer, M.C.: Induction of multiscale temporal structure. In: Advances in Neural Information Processing Systems, pp. 275–275 (1993)Google Scholar
  17. 17.
    Nierhaus, G.: Algorithmic Composition: Paradigms of Automated Music Generation. Springer Science & Business Media, Verlag (2009)CrossRefMATHGoogle Scholar
  18. 18.
    Sigtia, S., Benetos, E., Cherla, S., Weyde, T., Garcez, A.S.d., Dixon, S.: An RNN-based music language model for improving automatic music transcription. In: International Society for Music Information Retrieval Conference (ISMIR) (2014)Google Scholar
  19. 19.
    Sutskever, I.: Training recurrent neural networks. Ph.D. thesis, University of Toronto (2013)Google Scholar
  20. 20.
    Theano Development Team: Theano: A Python framework for fast computation of mathematical expressions. arXiv e-prints abs/1605.02688, May 2016. http://arxiv.org/abs/1605.02688
  21. 21.
    Tieleman, T., Hinton, G.: Lecture 6.5-rmsprop: Divide the gradient by a running average of its recent magnitude. In: COURSERA: Neural Networks for Machine Learning 4 (2012)Google Scholar
  22. 22.
    Todd, P.M.: A connectionist approach to algorithmic composition. Comput. Music J. 13(4), 27–43 (1989)CrossRefGoogle Scholar
  23. 23.
    Towns, J., Cockerill, T., Dahan, M., Foster, I., Gaither, K., Grimshaw, A., Hazlewood, V., Lathrop, S., Lifka, D., Peterson, G.D., et al.: XSEDE: accelerating scientific discovery. Comput. Sci. Eng. 16(5), 62–74 (2014)CrossRefGoogle Scholar
  24. 24.
    Vezhnevets, A., Mnih, V., Osindero, S., Graves, A., Vinyals, O., Agapiou, J., et al.: Strategic attentive writer for learning macro-actions. In: Advances in Neural Information Processing Systems, pp. 3486–3494 (2016)Google Scholar
  25. 25.
    Vohra, R., Goel, K., Sahoo, J.: Modeling temporal dependencies in data using a DBN-LSTM. In: IEEE International Conference on Data Science and Advanced Analytics (DSAA), 2015. 36678 2015, pp. 1–4 (2015)Google Scholar

Copyright information

© Springer International Publishing AG 2017

Authors and Affiliations

  1. 1.Harvey Mudd CollegeClaremontUSA

Personalised recommendations