Automatic acoustic detection of birds through deep learning : the first bird audio detection challenge

Stowell, D, Wood, M ORCID: https://orcid.org/0000-0002-0635-2387, Pamuła, H, Stylianou, Y and Glotin, H 2018, 'Automatic acoustic detection of birds through deep learning : the first bird audio detection challenge' , Methods in Ecology and Evolution, 10 (3) , pp. 368-380.

[img]
Preview
PDF - Published Version
Available under License Creative Commons Attribution 4.0.

Download (1MB) | Preview
[img] PDF - Accepted Version
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

  1. Assessing the presence and abundance of birds is important for monitoring specific species as well as overall ecosystem health. Many birds are most readily detected by their sounds, and thus passive acoustic monitoring is highly appropriate. Yet acoustic monitoring is often held back by practical limitations such as the need for manual configuration, reliance on example sound libraries, low accuracy, low robustness, and limited ability to generalise to novel acoustic conditions.
  2. Here we report outcomes from a collaborative data challenge. We present new acoustic monitoring datasets, summarise the machine learning techniques proposed by challenge teams, conduct detailed performance evaluation, and discuss how such approaches to detection can be integrated into remote monitoring projects.
  3. Multiple methods were able to attain performance of around 88% AUC (area under the ROC curve), much higher performance than previous general‐purpose methods.
  4. With modern machine learning including deep learning, general‐purpose acoustic bird detection can achieve very high retrieval rates in remote monitoring data ̶ with no manual recalibration, and no pre‐training of the detector for the target species or the acoustic conditions in the target environment.

Item Type: Article
Schools: Schools > School of Environment and Life Sciences > Ecosystems and Environment Research Centre
Journal or Publication Title: Methods in Ecology and Evolution
Publisher: Wiley
ISSN: 2041-210X
Related URLs:
Funders: Natural Environment Research Council (NERC), Environment Agency, Radioactiove Waste Management Ltd
Depositing User: Prof Mike Wood
Date Deposited: 22 Oct 2018 11:59
Last Modified: 12 Sep 2019 14:30
URI: http://usir.salford.ac.uk/id/eprint/48716

Actions (login required)

Edit record (repository staff only) Edit record (repository staff only)

Downloads

Downloads per month over past year