Automatic identification of individual killer whales

Brown, Judith C.; Smaragdis, Paris; Nousek‐McGregor, Anna E.

doi:10.1121/1.3462232

Cited by 15 publications

(8 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Third, MFCCs show high accuracy, stability, and repeatability (Cheng et al 2010). And finally, MFCCs are now also increasingly applied to caller recognition of animals such as elephants (Clemins et al 2005), passerine birds (Trawicki et al 2005;Fox 2008;Cheng et al 2010), toothed whales (Brown et al 2010), and blue monkeys (Mielke and Zuberbühler 2013).…”

Section: Introductionmentioning

confidence: 99%

Who shall I say is calling? Validation of a caller recognition procedure in Bornean flanged male orangutan (Pongo pygmaeus wurmbii) long calls

et al. 2016

View full text Add to dashboard Cite

Acoustic individual discrimination has been demonstrated for a wide range of animal taxa. However, there has been far less scientific effort to demonstrate the effectiveness of automatic individual identification, which could greatly facilitate research, especially when data are collected via an acoustic localization system (ALS). In this study, we examine the accuracy of acoustic caller recognition in long calls (LCs) emitted by Bornean male orangutans (Pongo pygmaeus wurmbii) derived from two data-sets: the first consists of high-quality recordings taken during individual focal follows (N = 224 LCs by 14 males) and the second consists of LC recordings with variable microphone-caller distances stemming from ALS (N = 123 LCs by 10 males). The LC is a long-distance vocalization. We therefore expect that even the low-quality test-set should yield caller recognition results significantly better than by chance. Automatic individual identification was accomplished using software originally developed for human speaker recognition (i.e. the MSR identity toolbox). We obtained a 93.3% correct identification rate with high-quality recordings, and 72.23% with recordings stemming from the ALS with variable microphone-caller distances (20-420 m). These results show that automatic individual identification is possible even though the accuracy declines compared with the results of high-quality recordings due to severe signal degradations (e.g. sound attenuation, environmental noise contamination, and echo interference) with increasing distance. We therefore suggest that acoustic individual identification with speaker recognition software can be a valuable tool to apply to data obtained through an ALS, thereby facilitating field research on vocal communication.

show abstract

Section: Introductionmentioning

confidence: 99%

Who shall I say is calling? Validation of a caller recognition procedure in Bornean flanged male orangutan (Pongo pygmaeus wurmbii) long calls

et al. 2016

View full text Add to dashboard Cite

show abstract

“…Gaussian Mixture Models (GMMs) and Support Vector Machine (SVM) algorithms have been used to construct cetacean species detectors to discriminate between signals produced by Blainville’s beaked whales ( Mesoplodon densirostris ), short-finned pilot whales ( Globicephala macrorhynchus ), and Risso’s dolphins ( Grampus griseus ) 20 ; and similar computational techniques have been implemented in an effort to estimate sperm whale size distributions 21 . It has been demonstrated that a radial basis function network could effectively distinguish between six individual sperm whales 22 ; and similarly, Hidden Markov Models (HMMs) and GMMs have performed the automatic identification of individual killer whales ( Orcinus orca ) 23 . In addition, ANNs have been constructed to classify the bioacoustic signals of killer whales based on call type, individual whale identity, and community dialect 24,25 .…”

Section: Introductionmentioning

confidence: 99%

Deep Machine Learning Techniques for the Detection and Classification of Sperm Whale Bioacoustics

Bermant

Bronstein

Wood

et al. 2019

Sci Rep

View full text Add to dashboard Cite

We implemented Machine Learning (ML) techniques to advance the study of sperm whale ( Physeter macrocephalus ) bioacoustics. This entailed employing Convolutional Neural Networks (CNNs) to construct an echolocation click detector designed to classify spectrograms generated from sperm whale acoustic data according to the presence or absence of a click. The click detector achieved 99.5% accuracy in classifying 650 spectrograms. The successful application of CNNs to clicks reveals the potential of future studies to train CNN-based architectures to extract finer-scale details from cetacean spectrograms. Long short-term memory and gated recurrent unit recurrent neural networks were trained to perform classification tasks, including (1) “coda type classification” where we obtained 97.5% accuracy in categorizing 23 coda types from a Dominica dataset containing 8,719 codas and 93.6% accuracy in categorizing 43 coda types from an Eastern Tropical Pacific (ETP) dataset with 16,995 codas; (2) “vocal clan classification” where we obtained 95.3% accuracy for two clan classes from Dominica and 93.1% for four ETP clan types; and (3) “individual whale identification” where we obtained 99.4% accuracy using two Dominica sperm whales. These results demonstrate the feasibility of applying ML to sperm whale bioacoustics and establish the validity of constructing neural networks to learn meaningful representations of whale vocalizations.

show abstract

“…calls, vocalization, echolocation trains and clicks) in a specific type of noise [33]. A plethora of tools from pattern to voice recognition has been developed to build signal detectors relative to a single species or group of affiliate species [35,[40][41][42][43]. Automatic detection of biological signals at higher organization levels (communities) has been rarely pursued [12,44] owing to difficulties in dealing with other sources of sounds (anthropogenic or geophysical noise).…”

Section: (A) Bioacoustic and Ecoacoustic Indexesmentioning

confidence: 99%

Acoustic indexes for marine biodiversity trends and ecosystem health

Pieretti

Danovaro

2020

Phil. Trans. R. Soc. B

View full text Add to dashboard Cite

Acoustic approaches have been recently proposed to investigate critical ecological issues, such as biodiversity loss and different typologies of impacts, including climate change. However, the extensive use of acoustic monitoring is hampered by the lack of algorithms enabling the discrimination among different sound sources (e.g. geophysical, anthropogenic and biological). Eco- and bioacoustic indexes have been applied to provide non-invasive information on the temporal and spatial patterns of marine biodiversity and on the anthropogenic impact on marine life. Here, we review the potential of acoustic tools in expanding the monitoring of marine ecosystems from a current three-dimensional perception to a four-dimensional dimension. We also explore the use of acoustic indexes, mostly developed in terrestrial ecology, for the investigation of different marine ecosystems. Their appraisal, strengths and limits, and potential for future investigations in the biological exploration of the oceans are also discussed. This article is part of the theme issue ‘Integrative research perspectives on marine conservation’.

show abstract

Automatic identification of individual killer whales

Cited by 15 publications

References 12 publications

Who shall I say is calling? Validation of a caller recognition procedure in Bornean flanged male orangutan (Pongo pygmaeus wurmbii) long calls

Who shall I say is calling? Validation of a caller recognition procedure in Bornean flanged male orangutan (Pongo pygmaeus wurmbii) long calls

Deep Machine Learning Techniques for the Detection and Classification of Sperm Whale Bioacoustics

Acoustic indexes for marine biodiversity trends and ecosystem health

Contact Info

Product

Resources

About