Slow-Fast Auditory Streams For Audio Recognition

Code

• github.com/ekazakos/auditory-slow-fast
  OfficialIn paperpytorch★ 73
• github.com/porcelluscavia/audio-model
  pytorch★ 3

Abstract

We propose a two-stream convolutional network for audio recognition, that operates on time-frequency spectrogram inputs. Following similar success in visual recognition, we learn Slow-Fast auditory streams with separable convolutions and multi-level lateral connections. The Slow pathway has high channel capacity while the Fast pathway operates at a fine-grained temporal resolution. We showcase the importance of our two-stream proposal on two diverse datasets: VGG-Sound and EPIC-KITCHENS-100, and achieve state-of-the-art results on both.

Tasks

Benchmark Results

Reproductions