Passive acoustic monitoring (PAM) is a useful technique for monitoring marine mammals. However, the quantity of data collected through PAM systems makes automated algorithms for detecting and classifying sounds essential. Deep learning algorithms have shown great promise in recent years, but their performance is limited by the lack of sufficient amounts of annotated data for training the algorithms. This work investigates the benefit of augmenting training datasets with synthetically generated samples when training a deep neural network for the classification of North Atlantic right whale (Eubalaena glacialis) upcalls. We apply two recently proposed augmentation techniques, SpecAugment and Mixup, and show that they improve the performance of our model considerably. The precision is increased from 86% to 90%, while the recall is increased from 88% to 93%. Finally, we demonstrate that these two methods yield a significant improvement in performance in a scenario of data scarcity, where few training samples are available. This demonstrates that data augmentation can reduce the annotation effort required to achieve a desirable performance threshold.
Skip Nav Destination
Article navigation
April 2021
April 12 2021
Data augmentation for the classification of North Atlantic right whales upcallsa)
Special Collection:
Machine Learning in Acoustics
Bruno Padovese
;
Bruno Padovese
b)
Institute for Big Data Analytics, Dalhousie University
, Halifax, Nova Scotia, B3H 4R2, Canada
Search for other works by this author on:
Fabio Frazao;
Fabio Frazao
Institute for Big Data Analytics, Dalhousie University
, Halifax, Nova Scotia, B3H 4R2, Canada
Search for other works by this author on:
Oliver S. Kirsebom
;
Oliver S. Kirsebom
c)
Institute for Big Data Analytics, Dalhousie University
, Halifax, Nova Scotia, B3H 4R2, Canada
Search for other works by this author on:
Stan Matwin
Stan Matwin
Institute for Big Data Analytics, Dalhousie University
, Halifax, Nova Scotia, B3H 4R2, Canada
Search for other works by this author on:
a)
This paper is part of a special issue on Machine Learning in Acoustics.
b)
Electronic mail: bpadovese@dal.ca, ORCID: 0000-0002-9879-9900.
c)
ORCID: 0000-0001-5843-7465.
J. Acoust. Soc. Am. 149, 2520–2530 (2021)
Article history
Received:
November 19 2020
Accepted:
March 23 2021
Citation
Bruno Padovese, Fabio Frazao, Oliver S. Kirsebom, Stan Matwin; Data augmentation for the classification of North Atlantic right whales upcalls. J. Acoust. Soc. Am. 1 April 2021; 149 (4): 2520–2530. https://doi.org/10.1121/10.0004258
Download citation file:
Sign in
Don't already have an account? Register
Sign In
You could not be signed in. Please check your credentials and make sure you have an active account and try again.
Sign in via your Institution
Sign in via your InstitutionPay-Per-View Access
$40.00