Corpus ID: 220041617

Incorporating Music Knowledge in Continual Dataset Augmentation for Music Generation

@article{Liu2020IncorporatingMK,
  title={Incorporating Music Knowledge in Continual Dataset Augmentation for Music Generation},
  author={Alisa Liu and A. Fang and Gaetan Hadjeres and P. Seetharaman and B. Pardo},
  journal={ArXiv},
  year={2020},
  volume={abs/2006.13331}
}
  • Alisa Liu, A. Fang, +2 authors B. Pardo
  • Published 2020
  • Computer Science, Engineering, Mathematics
  • ArXiv
  • Deep learning has rapidly become the state-of-the-art approach for music generation. However, training a deep model typically requires a large training set, which is often not available for specific musical styles. In this paper, we present augmentative generation (Aug-Gen), a method of dataset augmentation for any music generation system trained on a resource-constrained domain. The key intuition of this method is that the training data for a generative system can be augmented by examples the… CONTINUE READING

    Figures and Topics from this paper

    Explore Further: Topics Discussed in This Paper

    References

    SHOWING 1-10 OF 11 REFERENCES
    DeepBach: a Steerable Model for Bach Chorales Generation
    • 226
    • PDF
    Deep Learning Techniques for Music Generation - A Survey
    • 154
    • PDF
    An Improved Relative Self-Attention Mechanism for Transformer with Application to Music Generation
    • 36
    Not Enough Data? Deep Learning to the Rescue!
    • 17
    • PDF
    Attention is All you Need
    • 15,890
    • PDF
    Active Generative Adversarial Network for Image Classification
    • 7
    • PDF
    Music21: A Toolkit for Computer-Aided Musicology and Symbolic Music Data
    • 234
    • PDF
    A grading function for four-part chorales in the style of j.s
    • bach. Under review,
    • 2020
    A grading function for four-part chorales in the style of j.s. bach. Under review
    • 2020