Slow-Fast Auditory Streams for Audio Recognition (2021-03-05T00:00:00.000000Z)

TL;DR

This work learns Slow-Fast auditory streams with separable convolutions and multi-level lateral connections in a two-stream convolutional network for audio recognition, that operates on time-frequency spectrogram inputs.

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.

Authors

Andrew Zisserman

62 papers

D. Damen

12 papers

E. Kazakos

5 papers

TL;DR

Abstract

Authors

References29 items

Device-Robust Acoustic Scene Classification Based on Two-Stage Categorization and Data Augmentation

Acoustic Scene Classification Using Deep Residual Networks with Late Fusion of Separated High and Low Frequency Paths

Time-Frequency Feature Decomposition Based on Sound Duration for Acoustic Scene Classification

Deep Audio Priors Emerge From Harmonic Convolutional Networks

Vggsound: A Large-Scale Audio-Visual Dataset

Audiovisual SlowFast Networks for Video Recognition

Spectro-Temporal Processing in a Two-Stream Computational Model of Auditory Cortex

State-of-the-Art Speech Recognition Using Multi-Stream Self-Attention with Dilated 1D Convolutions

Integrating the Data Augmentation Scheme with Various Classifiers for Acoustic Scene Modeling

The Receptive Field as a Regularizer in Deep Convolutional Neural Networks for Acoustic Scene Classification

SpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition

Environment Sound Classification Using a Two-Stream CNN Based on Decision-Level Fusion

Multi-stream Network With Temporal Attention For Environmental Sound Classification

SlowFast Networks for Video Recognition

Acoustic Features Fusion using Attentive Multi-channel Deep Architecture

A Closer Look at Spatiotemporal Convolutions for Action Recognition

Comparison of Time-Frequency Representations for Environmental Sound Classification using Convolutional Neural Networks

Timbre analysis of music audio signals with convolutional neural networks

Convolutional neural networks for acoustic scene classiﬁcation

SoundNet: Learning Sound Representations from Unlabeled Video

CNN architectures for large-scale audio classification

Deep Convolutional Neural Networks and Data Augmentation for Environmental Sound Classification

Temporal Segment Networks: Towards Good Practices for Deep Action Recognition

Deep Residual Learning for Image Recognition

Encoding of Natural Sounds at Multiple Spectral and Temporal Resolutions in the Human Auditory Cortex

“Rescaling egocentric vision,”

Designing Acoustic Scene Classification Models with CNN Variants Technical

Acoustic scene classification using multi-scale features

“Acoustic scene classiﬁ-cation with fully convolutional neural networks and I-vectors,”

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names