A trainable comb filter layer extracts harmonic features from raw audio with one parameter per channel, matching convolutional frontends on three audio tasks with orders of magnitude lower cost.
Robust speech recognition via large-scale weak supervision,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.SD 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Combolutional Neural Networks
A trainable comb filter layer extracts harmonic features from raw audio with one parameter per channel, matching convolutional frontends on three audio tasks with orders of magnitude lower cost.