ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning
Large bioacoustic archives of wild animals are an important source to identify reappearing communication patterns, which can then be related to recurring behavioral patterns to advance the current understanding of intra-specific communication of non-human animals. A main challenge remains that most...
Main Authors: | , , , , , , , |
---|---|
Format: | Text |
Language: | unknown |
Published: |
Freie Universität Berlin
2019
|
Subjects: | |
Online Access: | https://dx.doi.org/10.17169/refubium-4018 https://refubium.fu-berlin.de/handle/fub188/25315 |
id |
ftdatacite:10.17169/refubium-4018 |
---|---|
record_format |
openpolar |
spelling |
ftdatacite:10.17169/refubium-4018 2023-05-15T17:03:23+02:00 ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning Bergler, Christian Schröter, Hendrik Cheng, Rachael Xi Barth, Volker Weber, Michael Nöth, Elmar Hofer, Heribert Maier, Andreas 2019 https://dx.doi.org/10.17169/refubium-4018 https://refubium.fu-berlin.de/handle/fub188/25315 unknown Freie Universität Berlin https://doi.org/10.1038/s41598-019-47335-w https://dx.doi.org/10.1038/s41598-019-47335-w https://doi.org/10.1038/s41598-019-47335-w Creative Commons Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/legalcode cc-by-4.0 CC-BY animal behaviour behavioural ecology 500 Naturwissenschaften und Mathematik590 Tiere Zoologie599 Mammalia Säugetiere Text article-journal Wissenschaftlicher Artikel ScholarlyArticle 2019 ftdatacite https://doi.org/10.17169/refubium-4018 https://doi.org/10.1038/s41598-019-47335-w 2021-11-05T12:55:41Z Large bioacoustic archives of wild animals are an important source to identify reappearing communication patterns, which can then be related to recurring behavioral patterns to advance the current understanding of intra-specific communication of non-human animals. A main challenge remains that most large-scale bioacoustic archives contain only a small percentage of animal vocalizations and a large amount of environmental noise, which makes it extremely difficult to manually retrieve sufficient vocalizations for further analysis – particularly important for species with advanced social systems and complex vocalizations. In this study deep neural networks were trained on 11,509 killer whale (Orcinus orca) signals and 34,848 noise segments. The resulting toolkit ORCA-SPOT was tested on a large-scale bioacoustic repository – the Orchive – comprising roughly 19,000 hours of killer whale underwater recordings. An automated segmentation of the entire Orchive recordings (about 2.2 years) took approximately 8 days. It achieved a time-based precision or positive-predictive-value (PPV) of 93.2% and an area-under-the-curve (AUC) of 0.9523. This approach enables an automated annotation procedure of large bioacoustics databases to extract killer whale sounds, which are essential for subsequent identification of significant communication patterns. The code will be publicly available in October 2019 to support the application of deep learning to bioaoucstic research. ORCA-SPOT can be adapted to other animal species. Text Killer Whale Orca Orcinus orca Killer whale DataCite Metadata Store (German National Library of Science and Technology) |
institution |
Open Polar |
collection |
DataCite Metadata Store (German National Library of Science and Technology) |
op_collection_id |
ftdatacite |
language |
unknown |
topic |
animal behaviour behavioural ecology 500 Naturwissenschaften und Mathematik590 Tiere Zoologie599 Mammalia Säugetiere |
spellingShingle |
animal behaviour behavioural ecology 500 Naturwissenschaften und Mathematik590 Tiere Zoologie599 Mammalia Säugetiere Bergler, Christian Schröter, Hendrik Cheng, Rachael Xi Barth, Volker Weber, Michael Nöth, Elmar Hofer, Heribert Maier, Andreas ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning |
topic_facet |
animal behaviour behavioural ecology 500 Naturwissenschaften und Mathematik590 Tiere Zoologie599 Mammalia Säugetiere |
description |
Large bioacoustic archives of wild animals are an important source to identify reappearing communication patterns, which can then be related to recurring behavioral patterns to advance the current understanding of intra-specific communication of non-human animals. A main challenge remains that most large-scale bioacoustic archives contain only a small percentage of animal vocalizations and a large amount of environmental noise, which makes it extremely difficult to manually retrieve sufficient vocalizations for further analysis – particularly important for species with advanced social systems and complex vocalizations. In this study deep neural networks were trained on 11,509 killer whale (Orcinus orca) signals and 34,848 noise segments. The resulting toolkit ORCA-SPOT was tested on a large-scale bioacoustic repository – the Orchive – comprising roughly 19,000 hours of killer whale underwater recordings. An automated segmentation of the entire Orchive recordings (about 2.2 years) took approximately 8 days. It achieved a time-based precision or positive-predictive-value (PPV) of 93.2% and an area-under-the-curve (AUC) of 0.9523. This approach enables an automated annotation procedure of large bioacoustics databases to extract killer whale sounds, which are essential for subsequent identification of significant communication patterns. The code will be publicly available in October 2019 to support the application of deep learning to bioaoucstic research. ORCA-SPOT can be adapted to other animal species. |
format |
Text |
author |
Bergler, Christian Schröter, Hendrik Cheng, Rachael Xi Barth, Volker Weber, Michael Nöth, Elmar Hofer, Heribert Maier, Andreas |
author_facet |
Bergler, Christian Schröter, Hendrik Cheng, Rachael Xi Barth, Volker Weber, Michael Nöth, Elmar Hofer, Heribert Maier, Andreas |
author_sort |
Bergler, Christian |
title |
ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning |
title_short |
ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning |
title_full |
ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning |
title_fullStr |
ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning |
title_full_unstemmed |
ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning |
title_sort |
orca-spot: an automatic killer whale sound detection toolkit using deep learning |
publisher |
Freie Universität Berlin |
publishDate |
2019 |
url |
https://dx.doi.org/10.17169/refubium-4018 https://refubium.fu-berlin.de/handle/fub188/25315 |
genre |
Killer Whale Orca Orcinus orca Killer whale |
genre_facet |
Killer Whale Orca Orcinus orca Killer whale |
op_relation |
https://doi.org/10.1038/s41598-019-47335-w https://dx.doi.org/10.1038/s41598-019-47335-w https://doi.org/10.1038/s41598-019-47335-w |
op_rights |
Creative Commons Attribution 4.0 International https://creativecommons.org/licenses/by/4.0/legalcode cc-by-4.0 |
op_rightsnorm |
CC-BY |
op_doi |
https://doi.org/10.17169/refubium-4018 https://doi.org/10.1038/s41598-019-47335-w |
_version_ |
1766057241766526976 |