Introduces an optimal transport loss for neural piano transcription to accommodate temporal misalignments, achieving state-of-the-art onset detection on the MAESTRO dataset via a harmonics-aware CRNN.
Scoring time intervals using non- hierarchical transformer for automatic piano transcription,
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
-
A Distribution Matching Approach to Neural Piano Transcription with Optimal Transport
Introduces an optimal transport loss for neural piano transcription to accommodate temporal misalignments, achieving state-of-the-art onset detection on the MAESTRO dataset via a harmonics-aware CRNN.