City Research Online

Abstract

In this study we explore the use of deep feedforward neural networks for voice separation in symbolic music representations. We experiment with different network architectures, varying the number and size of the hidden layers, and with dropout. We integrate two voice entry estimation heuristics that estimate the entry points of the individual voices in the polyphonic fabric into the models. These heuristics serve to reduce error propagation at the beginning of a piece, which, as we have shown in previous work, can seriously hamper model performance.

The models are evaluated on the 48 fugues from Johann Sebastian Bach’s The Well-Tempered Clavier and his 30 inventions—a dataset that we curated and make publicly available. We find that a model with two hidden layers yields the best results. Using more layers does not lead to a significant performance improvement. Furthermore, we find that our voice entry estimation heuristics are highly effective in the reduction of error propagation, improving performance significantly. Our best-performing model outperforms our previous models, where the difference is significant, and, depending on the evaluation metric, performs close to or better than the reported state of the art.

Publication Type:	Conference or Workshop Item (Paper)
Subjects:	B Philosophy. Psychology. Religion > BF Psychology M Music and Books on Music R Medicine > RC Internal medicine > RC0321 Neuroscience. Biological psychiatry. Neuropsychiatry
Departments:	School of Science & Technology > Computer Science

Export

Downloads