[3N1-IS-2d] Machine learning (4/5)

[3N1-IS-2d-04] MIDI note embedding with fastText model

〇Yingfeng Fu¹, Yusuke Tanimura^2,1, Hidemoto Nakada^2,1 (1. University of Tsukuba, 2. AIST)

キーワード：FastText, Word embedding, Music information retrieval

Distributed word representation greatly promoted research in NLP. Same as languages, MIDI music is constructed in the way of sequence, with a determined alphabet of notes and events. We proposed a way of training MIDI note embedding with an adaption of Facebook's fastText model. We then evaluate the model by word similarity, word analogy, and a classification task. The result shows that the adopted fastText model generalizes well in MIDI data and it’s promising to be used on future downstream tasks.