Large-scale unsupervised clustering of Orca vocalizations: a model for describing orca communication systems
Killer whales (Orcinus orca) can produce 3 types of signals: clicks, whistles and vocalizations. This study focuses on Orca vocalizations from northern Vancouver Island (Hanson Island) where the NGO Orcalab developed a multi-hydrophone recording station to study Orcas. The acoustic station is compos...
Main Authors: | , , , , , |
---|---|
Format: | Other/Unknown Material |
Language: | unknown |
Published: |
PeerJ
2019
|
Subjects: | |
Online Access: | http://dx.doi.org/10.7287/peerj.preprints.27979 https://peerj.com/preprints/27979.pdf https://peerj.com/preprints/27979.xml https://peerj.com/preprints/27979.html |
id |
crpeerj:10.7287/peerj.preprints.27979 |
---|---|
record_format |
openpolar |
spelling |
crpeerj:10.7287/peerj.preprints.27979 2024-06-02T08:12:43+00:00 Large-scale unsupervised clustering of Orca vocalizations: a model for describing orca communication systems Poupard, Marion Best, Paul Schlüter, Jan Symonds, Helena Spong, Paul Glotin, Hervé 2019 http://dx.doi.org/10.7287/peerj.preprints.27979 https://peerj.com/preprints/27979.pdf https://peerj.com/preprints/27979.xml https://peerj.com/preprints/27979.html unknown PeerJ https://creativecommons.org/licenses/by/4.0/ posted-content 2019 crpeerj https://doi.org/10.7287/peerj.preprints.27979 2024-05-07T14:14:00Z Killer whales (Orcinus orca) can produce 3 types of signals: clicks, whistles and vocalizations. This study focuses on Orca vocalizations from northern Vancouver Island (Hanson Island) where the NGO Orcalab developed a multi-hydrophone recording station to study Orcas. The acoustic station is composed of 5 hydrophones and extends over 50 km 2 of ocean. Since 2015 we are continuously streaming the hydrophone signals to our laboratory in Toulon, France, yielding nearly 50 TB of synchronous multichannel recordings. In previous work, we trained a Convolutional Neural Network (CNN) to detect Orca vocalizations, using transfer learning from a bird activity dataset. Here, for each detected vocalization, we estimate the pitch contour (fundamental frequency). Finally, we cluster vocalizations by features describing the pitch contour. While preliminary, our results demonstrate a possible route towards automatic Orca call type classification. Furthermore, they can be linked to the presence of particular Orca pods in the area according to the classification of their call types. A large-scale call type classification would allow new insights on phonotactics and ethoacoustics of endangered Orca populations in the face of increasing anthropic pressure. Other/Unknown Material Orca Orcinus orca PeerJ Publishing |
institution |
Open Polar |
collection |
PeerJ Publishing |
op_collection_id |
crpeerj |
language |
unknown |
description |
Killer whales (Orcinus orca) can produce 3 types of signals: clicks, whistles and vocalizations. This study focuses on Orca vocalizations from northern Vancouver Island (Hanson Island) where the NGO Orcalab developed a multi-hydrophone recording station to study Orcas. The acoustic station is composed of 5 hydrophones and extends over 50 km 2 of ocean. Since 2015 we are continuously streaming the hydrophone signals to our laboratory in Toulon, France, yielding nearly 50 TB of synchronous multichannel recordings. In previous work, we trained a Convolutional Neural Network (CNN) to detect Orca vocalizations, using transfer learning from a bird activity dataset. Here, for each detected vocalization, we estimate the pitch contour (fundamental frequency). Finally, we cluster vocalizations by features describing the pitch contour. While preliminary, our results demonstrate a possible route towards automatic Orca call type classification. Furthermore, they can be linked to the presence of particular Orca pods in the area according to the classification of their call types. A large-scale call type classification would allow new insights on phonotactics and ethoacoustics of endangered Orca populations in the face of increasing anthropic pressure. |
format |
Other/Unknown Material |
author |
Poupard, Marion Best, Paul Schlüter, Jan Symonds, Helena Spong, Paul Glotin, Hervé |
spellingShingle |
Poupard, Marion Best, Paul Schlüter, Jan Symonds, Helena Spong, Paul Glotin, Hervé Large-scale unsupervised clustering of Orca vocalizations: a model for describing orca communication systems |
author_facet |
Poupard, Marion Best, Paul Schlüter, Jan Symonds, Helena Spong, Paul Glotin, Hervé |
author_sort |
Poupard, Marion |
title |
Large-scale unsupervised clustering of Orca vocalizations: a model for describing orca communication systems |
title_short |
Large-scale unsupervised clustering of Orca vocalizations: a model for describing orca communication systems |
title_full |
Large-scale unsupervised clustering of Orca vocalizations: a model for describing orca communication systems |
title_fullStr |
Large-scale unsupervised clustering of Orca vocalizations: a model for describing orca communication systems |
title_full_unstemmed |
Large-scale unsupervised clustering of Orca vocalizations: a model for describing orca communication systems |
title_sort |
large-scale unsupervised clustering of orca vocalizations: a model for describing orca communication systems |
publisher |
PeerJ |
publishDate |
2019 |
url |
http://dx.doi.org/10.7287/peerj.preprints.27979 https://peerj.com/preprints/27979.pdf https://peerj.com/preprints/27979.xml https://peerj.com/preprints/27979.html |
genre |
Orca Orcinus orca |
genre_facet |
Orca Orcinus orca |
op_rights |
https://creativecommons.org/licenses/by/4.0/ |
op_doi |
https://doi.org/10.7287/peerj.preprints.27979 |
_version_ |
1800759249247141888 |