Corpus ID: 9651443

Song From PI: A Musically Plausible Network for Pop Music Generation

@article{Chu2017SongFP,
  title={Song From PI: A Musically Plausible Network for Pop Music Generation},
  author={Hang Chu and R. Urtasun and S. Fidler},
  journal={ArXiv},
  year={2017},
  volume={abs/1611.03477}
}
  • Hang Chu, R. Urtasun, S. Fidler
  • Published 2017
  • Computer Science
  • ArXiv
  • We present a novel framework for generating pop music. Our model is a hierarchical Recurrent Neural Network, where the layers and the structure of the hierarchy encode our prior knowledge about how pop music is composed. In particular, the bottom layers generate the melody, while the higher levels produce the drums and chords. We conduct several human studies that show strong preference of our generated music over that produced by the recent method by Google. We additionally show two… CONTINUE READING
    74 Citations

    Figures, Tables, and Topics from this paper.

    Generating Nontrivial Melodies for Music as a Service
    • 9
    • PDF
    DeepJ: Style-Specific Music Generation
    • 24
    • PDF
    BandNet: A Neural Network-based, Multi-Instrument Beatles-Style MIDI Music Composition Machine
    • 2
    • PDF
    Melody Generation for Pop Music via Word Representation of Musical Properties
    • 6
    • Highly Influenced
    • PDF
    Neural Melody Composition from Lyrics
    • 8
    • PDF
    Polyphonic Music Composition with LSTM Neural Networks and Reinforcement Learning
    • 1
    • PDF
    Foley Music: Learning to Generate Music from Videos
    • 5
    • PDF
    Deep Composer: Deep Neural Hashing And Retrieval Approach To Automatic Music Generation
    A SeqGAN for Polyphonic Music Generation
    • 11
    • PDF

    References

    SHOWING 1-10 OF 29 REFERENCES
    Neural Network Music Composition by Prediction: Exploring the Benefits of Psychoacoustic Constraints and Multi-scale Processing
    • M. Mozer
    • Computer Science, Engineering
    • Connect. Sci.
    • 1994
    • 205
    • Highly Influential
    • PDF
    Deep Learning for Music
    • 47
    • PDF
    Modeling Temporal Dependencies in High-Dimensional Sequences: Application to Polyphonic Music Generation and Transcription
    • 514
    • Highly Influential
    • PDF
    Improving algorithmic music composition with machine learning
    • 9
    • PDF
    A First Look at Music Composition using LSTM Recurrent Neural Networks
    • 169
    • Highly Influential
    • PDF
    Creating melodies with evolving recurrent neural networks
    • C.-C.J. Chen, R. Miikkulainen
    • Computer Science
    • IJCNN'01. International Joint Conference on Neural Networks. Proceedings (Cat. No.01CH37222)
    • 2001
    • 85
    • PDF
    Automatic Music Generation and Machine Learning Based Evaluation
    • 2
    Automatic Real-Time Music Generation for Games
    • 5
    Modeling the Perception of Tonal Structure with Neural Nets
    • 106
    • PDF
    A Neural Algorithm of Artistic Style
    • 1,307
    • PDF