Conference item icon

Conference item

Slow-fast auditory streams for audio recognition

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.
Publication status:
Published
Peer review status:
Peer reviewed

Actions


Access Document


Files:
Publisher copy:
10.1109/ICASSP39728.2021.9413376

Authors


More by this author
Institution:
University of Oxford
Division:
MPLS
Department:
Engineering Science
Oxford college:
Brasenose College
Role:
Author
ORCID:
0000-0002-8945-8573
Publisher:
IEEE Publisher's website
Host title:
ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Pages:
855-859
Publication date:
2021-05-13
Acceptance date:
2021-01-29
Event title:
2021 IEEE International Conference on Acoustics, Speech and Signal Processing
Event location:
Toronto, Ontario, Canada
Event website:
https://www.2021.ieeeicassp.org/
Event start date:
2021-06-06
Event end date:
2021-06-11
DOI:
EISSN:
2379-190X
ISSN:
1520-6149
EISBN:
978-1-7281-7605-5
ISBN:
978-1-7281-7606-2
Language:
English
Keywords:
Pubs id:
1212419
Local pid:
pubs:1212419
Deposit date:
2022-01-19

Terms of use


Views and Downloads






If you are the owner of this record, you can report an update to it here: Report update to this record

TO TOP