English

BumbleBee: A Transformer for Music

Sound 2021-07-09 v1 Machine Learning Audio and Speech Processing

Abstract

We will introduce BumbleBee, a transformer model that will generate MIDI music data . We will tackle the issue of transformers applied to long sequences by implementing a longformer generative model that uses dilating sliding windows to compute the attention layers. We will compare our results to that of the music transformer and Long-Short term memory (LSTM) to benchmark our results. This analysis will be performed using piano MIDI files, in particular , the JSB Chorales dataset that has already been used for other research works (Huang et al., 2018)

Keywords

Cite

@article{arxiv.2107.03443,
  title  = {BumbleBee: A Transformer for Music},
  author = {Lucas Fenaux and Maria Juliana Quintero},
  journal= {arXiv preprint arXiv:2107.03443},
  year   = {2021}
}

Comments

8 pages, 3 figures

R2 v1 2026-06-24T03:58:44.332Z