Frequency dynamic convolution: Frequency-adaptive pattern recognition for sound event detection

Hyeonuk nam, Seong-hu Kim, byeong-yun ko, Yong-Hwa Park

Abstract:

2D convolution is widely used in sound event detection (SED)
to recognize two dimensional time-frequency patterns of sound
events. However, 2D convolution enforces translation equivariance
on sound events along both time and frequency axis while
frequency is not shift-invariant dimension. In order to improve
physical consistency of 2D convolution on SED, we propose
frequency dynamic convolution which applies kernel that adapts
to frequency components of input. Frequency dynamic convolution
outperforms the baseline by 6.3% in DESED validation
dataset in terms of polyphonic sound detection score (PSDS).
It also significantly outperforms other pre-existing contentadaptive
methods on SED. In addition, by comparing class-wise
F1 scores of baseline and frequency dynamic convolution, we
showed that frequency dynamic convolution is especially more
effective for detection of non-stationary sound events with intricate
time-frequency patterns. From this result, we verified
that frequency dynamic convolution is superior in recognizing
frequency-dependent patterns.