A T5-adapted transformer quantizes performance MIDI rhythms using beat annotations, achieving 97.3% onset F1-score and 83.3% note value accuracy on the ASAP dataset while generalizing to unseen time signatures.
Finale version 27
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Transformer-Based Rhythm Quantization of Performance MIDI Using Beat Annotations
A T5-adapted transformer quantizes performance MIDI rhythms using beat annotations, achieving 97.3% onset F1-score and 83.3% note value accuracy on the ASAP dataset while generalizing to unseen time signatures.