11 citations · 15 across the 10 of their papers we have counts for
1 paper · 2 filters
Kin Wai Cheuk, Ryosuke Sawata, Toshimitsu Uesaka +5
In this paper we propose a novel generative approach, DiffRoll, to tackle automatic music transcription (AMT). Instead of treating AMT as a discriminative task in which the model i…