Abstract
Marine mammals produce a wide variety of vocalizations. There is a growing need for robust automatic classification methods especially in noisy underwater environments in order to access large amounts of bioacoustic signals and to replace tedious and error prone human perceptual classification. In case of the northern resident killer whale (Orcinus orca), echolocation clicks, whistles, and pulsed calls make up its vocal repertoire. Pulsed calls are the most intensively studied type of vocalization. In this study we propose a hybrid call type classification approach outperforming our previous work on supervised call type classification consisting of two components: (1) deep representation learning of killer whale sounds by investigating various autoencoder architectures and data corpora and (2) subsequent supervised training of a ResNet18 call type classifier on a much smaller dataset by using the pre-trained representations. The best semi-supervised trained classification model achieved a test accuracy of 96% and a mean test accuracy of 94% outperforming our previous work by 7% points.
The authors would like to thank Helena Symonds and Paul Spong from Orcalab, and Steven Ness, formerly UVIC, for giving us permission to use the raw data and annotations from orcalab.org, and the Paul G. Allen Frontiers Group for their initial grant for the pilot research.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Bengio, Y., Courville, A.C., Vincent, P.: Representation learning: a review and new perspectives. IEEE Trans. Pattern Anal. Mach. Intell. 35, 1798–1828 (2013)
Bigg, M.A., Olesiuk, P.F., Ellis, G.M., Ford, J.K.B., Balcomb, K.C.: Organization and genealogy of resident killer whales (Orcinus orca) in the coastal waters of British Columbia and Washington State. Int. Whaling Comm. 12, 383–405 (1990)
Brown, J., Hodgins-Davis, A., Miller, P.: Classification of vocalizations of killer whales using dynamic time warping. JASA Express Lett. 119(3), 617–628 (2006)
Brown, J.C., Smaragdis, P.: Hidden Markov and Gaussian mixture models for automatic call classification. J. Acoust. Soc. Am. 125, 221–224 (2009)
Brown, J.C., Smaragdis, P., Nousek-McGregor, A.: Automatic identification of individual killer whales. J. Acoust. Soc. Am. 128, 93–98 (2010)
Deecke, V.B., Janik, V.M.: Automated categorization of bioacoustic signals: avoiding perceptual pitfalls. J. Acoust. Soc. Am. 119, 645–653 (2006)
Filatova, O.A., Samarra, F.I., Deecke, V.B., Ford, J.K., Miller, P.J., Yurk, H.: Cultural evolution of killer whale calls: background, mechanisms and consequences. Behaviour 152, 2001–2038 (2015)
Ford, J., Ellis, G., Balcomb, K.: Killer Whales: The Natural History and Genealogy of Orcinus Orca in British Columbia and Washington. UBC Press, Vancouver (2000)
Ford, J.K.B.: A catalogue of underwater calls produced by killer whales (Orcinus orca) in British Columbia. Canadian Data Report of Fisheries and Aquatic Science (633), p. 165 (1987)
Ford, J.K.B.: Acoustic behaviour of resident killer whales (Orcinus orca) off Vancouver Island, British Columbia. Can. J. Zool. 67, 727–745 (1989)
Ford, J.K.B.: Vocal traditions among resident killer whales (Orcinus orca) in coastal waters of British Columbia. Can. J. Zool. 69, 1454–1483 (1991)
Garland, E., Castellote, M., Berchok, C.: Beluga whale (Delphinapterus leucas) vocalizations and call classification from the eastern Beaufort sea population. J. Acoust. Soc. of Am. 137, 3054–3067 (2015)
Goodfellow, I., Bengio, Y., Courville, A.: Deep Learning. MIT Press, Cambridge (2016)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778 (2016)
Ivkovich, T., Filatova, O., Burdin, A., Sato, H., Hoyt, E.: The social organization of resident-type killer whales (Orcinus orca) in Avacha Gulf, Northwest Pacific, as revealed through association patterns and acoustic similarity. Mamm. Biol. 75, 198–210 (2010)
LeCun, Y., Bottou, L., Bengio, Y., Haffner, P.: Gradient-based learning applied to document recognition. Proc. IEEE 86, 2278–2324 (1998)
Mercado, E., Kuh, A.: Classification of humpback whale vocalizations using a self-organizing neural network. In: IEEE International Conference on Neural Networks - Conference Proceedings, pp. 1584–1589, June 1998
Miller, P., Bain, D.: Within-pod variation in the sound production of a pod of killer whales, Orcinus orca. Anim. Behav. 60, 617–628 (2000)
Ness, S.: The Orchive: a system for semi-automatic annotation and analysis of a large collection of bioacoustic recordings. Ph.D. thesis (2013)
ORCALAB: a whale research station on Hanson Island. http://orcalab.org. Accessed May 2019
Paszke, A., et al.: Automatic differentiation in PyTorch. In: NIPS 2017 Workshop, October 2017
Schröter, H., Nöth, E., Maier, A., Cheng, R., Barth, V., Bergler, C.: Segmentation, classification, and visualization of orca calls using deep learning. In: International Conference on Acoustics, Speech, and Signal Processing, Proceedings (ICASSP), May 2019
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Bergler, C. et al. (2019). Deep Representation Learning for Orca Call Type Classification. In: Ekštein, K. (eds) Text, Speech, and Dialogue. TSD 2019. Lecture Notes in Computer Science(), vol 11697. Springer, Cham. https://doi.org/10.1007/978-3-030-27947-9_23
Download citation
DOI: https://doi.org/10.1007/978-3-030-27947-9_23
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-27946-2
Online ISBN: 978-3-030-27947-9
eBook Packages: Computer ScienceComputer Science (R0)