SWIPE scores outperform a self-supervised neural pitch estimator (PESTO) on common benchmarks and, as an audio frontend, let a 647-parameter network match or beat the larger model.
The au- dio is re-synthesized from its f0 annotations, which means that the f0 annotations are perfect
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
dataset 1
citation-polarity summary
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1roles
dataset 1polarities
use dataset 1representative citing papers
citing papers explorer
-
Improving Neural Pitch Estimation with SWIPE Kernels
SWIPE scores outperform a self-supervised neural pitch estimator (PESTO) on common benchmarks and, as an audio frontend, let a 647-parameter network match or beat the larger model.