Transformer Neural Networks for Automated Rhythm Generation
Thomas Nuttall, Behzad Haki, and Sergi Jorda
Proceedings of the International Conference on New Interfaces for Musical Expression
- Year: 2021
- Location: Shanghai, China
- Article Number: 33
- DOI: 10.21428/92fbeb44.fe9a0d82 (Link to paper)
- PDF link
- Presentation Video
Abstract:
Recent applications of Transformer neural networks in the field of music have demonstrated their ability to effectively capture and emulate long-term dependencies characteristic of human notions of musicality and creative merit. We propose a novel approach to automated symbolic rhythm generation, where a Transformer-XL model trained on the Magenta Groove MIDI Dataset is used for the tasks of sequence generation and continuation. Hundreds of generations are evaluated using blind-listening tests to determine the extent to which the aspects of rhythm we understand to be valuable are learnt and reproduced. Our model is able to achieve a standard of rhythmic production comparable to human playing across arbitrarily long time periods and multiple playing styles.
Citation:
Thomas Nuttall, Behzad Haki, and Sergi Jorda. 2021. Transformer Neural Networks for Automated Rhythm Generation. Proceedings of the International Conference on New Interfaces for Musical Expression. DOI: 10.21428/92fbeb44.fe9a0d82BibTeX Entry:
@inproceedings{NIME21_33, abstract = {Recent applications of Transformer neural networks in the field of music have demonstrated their ability to effectively capture and emulate long-term dependencies characteristic of human notions of musicality and creative merit. We propose a novel approach to automated symbolic rhythm generation, where a Transformer-XL model trained on the Magenta Groove MIDI Dataset is used for the tasks of sequence generation and continuation. Hundreds of generations are evaluated using blind-listening tests to determine the extent to which the aspects of rhythm we understand to be valuable are learnt and reproduced. Our model is able to achieve a standard of rhythmic production comparable to human playing across arbitrarily long time periods and multiple playing styles.}, address = {Shanghai, China}, articleno = {33}, author = {Nuttall, Thomas and Haki, Behzad and Jorda, Sergi}, booktitle = {Proceedings of the International Conference on New Interfaces for Musical Expression}, doi = {10.21428/92fbeb44.fe9a0d82}, issn = {2220-4806}, month = {June}, presentation-video = {https://youtu.be/Ul9s8qSMUgU}, title = {Transformer Neural Networks for Automated Rhythm Generation}, url = {https://nime.pubpub.org/pub/8947fhly}, year = {2021} }