Skip to main content

Automatic Deep-Learning-Based Classification of Bottlenose Dolphin Signature Whistles

  • Living reference work entry
  • First Online:
The Effects of Noise on Aquatic Life

Abstract

Bottlenose dolphins (Tursiops truncatus) produce individually distinctive signature whistles that develop early in life and are used to recognize and maintain contact with conspecifics. Health assessments in Sarasota, Florida (USA), have provided a unique opportunity to record signature whistles of wild individuals of known age, sex, and matrilineal relationships. After 37 years, the Sarasota Dolphin recording library contains 930 recording sessions of 296 individual dolphins. Here, a deep convolutional neural network classifier was trained using a curated subset of 200 different signature whistles from each of 70 individual bottlenose dolphins. A MobileNetV2 trained on spectrogram data allocated signature whistles to the correct individual with an accuracy of 95.8%. To improve generalization to novel audio datasets, a data augmentation step was implemented that incorporated time stretching, pitch shifting, and mixing database whistles with natural background noise recordings. This augmented model achieved a classification accuracy of 94.5% for high signal-to-noise ratio (SNR) signature whistles, and 92.6% under more realistic conditions with signature whistles mixed with ambient noise at −6 to 12 dB SNR. These initial results are promising and suggest that automatic signature whistle classification techniques could enable acoustic monitoring of movements and habitat use of individual bottlenose dolphins at scale.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Institutional subscriptions

References

  • Barratclough A, Wells RS, Schwacke LH, Rowles TK, Gomez FM, Fauquier DA, Sweeney JC, Townsend FI, Hansen LJ, Zolman ES (2019) Health assessments of common bottlenose dolphins (Tursiops truncatus): past, present, and potential conservation applications. Front Vet Sci 6:444

    Article  PubMed  PubMed Central  Google Scholar 

  • Bergler C, Schröter H, Cheng RX, Barth V, Weber M, Nöth E, Hofer H, Maier A (2019) ORCA-SPOT: an automatic killer whale sound detection toolkit using deep learning. Sci Rep 9:10997

    Article  PubMed  PubMed Central  Google Scholar 

  • Buck JR, Tyack PL (1993) A quantitative measure of similarity for tursiops truncatus signature whistles. J Acoust Soc Am 94:2497–2506

    Article  PubMed  CAS  Google Scholar 

  • Buckstaff KC (2004) Effects of watercraft noise on the acoustic behaviour of bottlenose dolphins, Tursiops truncatus, in Sarasota Bay, Florida. Mar Mamm Sci 20:709–725

    Article  Google Scholar 

  • Caldwell MC, Caldwell DK (1965) Individualized whistle contours in bottle-nosed dolphins (Tursiops Truncatus). Nature 207:434

    Article  Google Scholar 

  • Caldwell MC, Caldwell DK, Tyack PL (1990) Review of the signature-whistle hypothesis for the Atlantic bottlenose dolphin. In: Leatherwood S, Reeves RR (eds) The bottlenose dolphin. Academic Press, New York, pp 199–234

    Chapter  Google Scholar 

  • Chereskin E, Connor RC, Friedman WR, Jensen FH, Allen SJ, Sørensen PM, Krützen M, King SL (2022) Allied male dolphins use vocal exchanges to “bond at a distance”. Curr Biol 32:1657–1663.e4

    Article  PubMed  CAS  Google Scholar 

  • Coffey KR, Marx RE, Neumaier JF (2019) DeepSqueak: a deep learning-based system for detection and analysis of ultrasonic vocalizations. Neuropsychopharmacology 44:859–868

    Article  PubMed  PubMed Central  Google Scholar 

  • Cook ML, Sayigh LS, Blum JE, Wells RS (2004) Signature–whistle production in undisturbed free–ranging bottlenose dolphins (Tursiops truncatus). Proc R Soc Lond Ser B Biol Sci 271:1043–1049

    Article  Google Scholar 

  • Deecke VB, Janik VM (2006) Automated categorization of bioacoustic signals: avoiding perceptual pitfalls. J Acoust Soc Am 119:645–653

    Article  PubMed  Google Scholar 

  • Gridley TG, Cockcroft VG, Hawkins ER, Lemon-Blewitt M, Morisaka T, Janik VM (2014) Signature whistles in free-ranging populations of Indo-Pacific bottlenose dolphins, Tursiops aduncus. Marine Mammal Science 30:512–527

    Google Scholar 

  • Janik VM (1997) Whistle matching in wild bottlenose dolphins. J Acoust Soc Am 101:3136–3136

    Article  Google Scholar 

  • Janik VM (2000) Whistle matching in wild bottlenose dolphins (Tursiops truncatus). Science 289:1355–1357

    Article  PubMed  CAS  Google Scholar 

  • Janik VM, Oswald J (2022) Combining neural network and sequence analyses to determine animal call repertoires. J Acoust Soc Am 151:A28–A28

    Article  Google Scholar 

  • Janik VM, Sayigh LS (2013) Communication in bottlenose dolphins: 50 years of signature whistle research. J Comp Physiol A 199:479–489

    Google Scholar 

  • Janik VM, Slater PJ (1998) Context-specific use suggests that bottlenose dolphin signature whistles are cohesion calls. Anim Behav 56:829–838

    Article  PubMed  CAS  Google Scholar 

  • Janik VM, Sayigh LS, Wells RS (2006) Signature whistle shape conveys identity information to bottlenose dolphins. Proc Natl Acad Sci U S A 103:8293–8297

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  • Janik VM, King SL, Sayigh LS, Wells RS (2013) Identifying signature whistles from recordings of groups of unrestrained bottlenose dolphins (Tursiops truncatus). Mar Mamm Sci 29:109–122

    Article  Google Scholar 

  • K. Lisa Yang Center for Conservation Bioacoustics (2022) Raven pro: interactive sound analysis software (version 1.6.3) [Computer software]. The Cornell Lab of Ornithology, Ithaca. Available from https://ravensoundsoftware.com/

  • Kershenbaum A, Sayigh LS, Janik VM (2013) The encoding of individual identity in dolphin signature whistles: how much information is needed? PLoS One 8:e77671

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  • King SL, Friedman WR, Allen SJ, Gerber L, Jensen FH, Wittwer S, Connor RC, Krutzen M (2018) Bottlenose dolphins retain individual vocal labels in multi-level alliances. Curr Biol 28:1993–1999.e3

    Article  PubMed  CAS  Google Scholar 

  • Longden EG, Elwen SH, McGovern B, James BS, Embling CB, Gridley T (2020) Mark–recapture of individually distinctive calls – a case study with signature whistles of bottlenose dolphins (Tursiops truncatus). J Mammal 101:1289–1301

    Article  Google Scholar 

  • Loughlin T, Cunningham L, Gales N, Wells RS, Boyd I (2010) Marking and capturing. In: Boyd I, Bowen D, Iverson S (eds) Marine mammal ecology and conservation: a handbook of techniques. Oxford University Press, Oxford

    Google Scholar 

  • NOAA Office of National Marine Sanctuaries and U.S Navy (2020) SanctSound raw passive acoustic data. NOAA National Centers for Environmental Information. https://doi.org/10.25921/saca-sp25. Access date 09 Dec 2021

  • Quick NJ, Janik VM (2012) Bottlenose dolphins exchange signature whistles when meeting at sea. Proc Biol Sci 279:2539–2545

    PubMed  PubMed Central  Google Scholar 

  • Quick NJ, Rendell LE, Janik VM (2008) A mobile acoustic localization system for the study of free-ranging dolphins during focal follows. Mar Mamm Sci 24:979–989

    Article  Google Scholar 

  • Rycyk AM, Tyson Moore RB, Wells RS, McHugh KA, Berens McCabe EJ, Mann DA (2020) Passive acoustic listening stations (PALS) show rapid onset of ecological effects of harmful algal blooms in real time. Sci Rep 10:1–12

    Article  Google Scholar 

  • Sandler M, Howard A, Zhu M, Zhmoginov A, Chen L-C (2018) Mobilenetv2: inverted residuals and linear bottlenecks. In: Proceedings of the IEEE conference on computer vision and pattern recognition, Salt Lake City, pp 4510–4520

    Google Scholar 

  • Sayigh LS, Tyack PL, Wells RS, Scott MD (1990) Signature whistles of free-ranging bottlenose dolphins Tursiops truncatus: stability and mother-offspring comparisons. Behav Ecol Sociobiol 26:247–260

    Article  Google Scholar 

  • Sayigh LS, Tyack PL, Wells RS, Solow AR, Scott MD, Irvine AB (1999) Individual recognition in wild bottlenose dolphins: a field test using playback experiments. Anim Behav 57:41–50

    Article  PubMed  CAS  Google Scholar 

  • Sayigh LS, Esch HC, Wells RS, Janik VM (2007) Facts about signature whistles of bottlenose dolphins, Tursiops truncatus. Anim Behav 74:1631–1642

    Article  Google Scholar 

  • Sayigh LS, Janik VM, Jensen FH, Scott MD, Tyack PL, Wells RS (2022) The Sarasota dolphin whistle database: a unique long-term resource for understanding dolphin communication. Front Mar Sci 9:923046

    Article  Google Scholar 

  • Shiu Y, Palmer KJ, Roch MA, Fleishman E, Liu X, Nosal E-M, Helble T, Cholewiak D, Gillespie D, Klinck H (2020) Deep neural networks for automated detection of marine mammal species. Sci Rep 10:607

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  • Watwood SL, Owen ECG, Tyack PL, Wells RS (2005) Signature whistle use by temporarily restrained and free-swimming bottlenose dolphins, Tursiops truncatus. Anim Behav 69:1373–1386

    Article  Google Scholar 

  • Wells RS (2009) Learning from nature: bottlenose dolphin care and husbandry. Zoo Biol 28:635–651

    Article  PubMed  Google Scholar 

  • Wells RS (2020) The Sarasota dolphin research program in 2020: celebrating 50 years of research, conservation, and education. Aquat Mamm 46:502–504

    Article  Google Scholar 

  • Wells RS, Rhinehart HL, Hansen LJ, Sweeney JC, Townsend FI, Stone R, Casper DR, Scott MD, Hohn AA, Rowles TK (2004) Bottlenose dolphins as marine ecosystem sentinels: developing a health monitoring system. EcoHealth 1:246–254

    Article  Google Scholar 

Download references

Acknowledgements

Many people have contributed to this multi-decadal effort, both in the field and in the lab. These include, but are not limited to: Blair Irvine and Michael Scott; staff, students, volunteers, and collaborators of the Sarasota Dolphin Research Program; students at UNCW and WHOI who have helped with data extraction, including Maia Austin, Gemma Bekki, Mandy Cook, Carter Esch, Kim Fleming, Gracie Gavazzi, Nicole el Haddad, Lynne Williams Hodge, Guen Jones, Kristi Kaleel, Jessica Maher, Larissa Michel, Lucia Snyderman, Campbell Van Horn, Jessica Veo, Nikki Vollmer, and Charles White. Financial support for data collection and for developing and curating the whistle database has come from the Protect Wild Dolphins fund at Harbor Branch Institute for Oceanography, Vulcan Machine Learning Center for Impact, Allen Institute for Artificial Intelligence, Link Foundation, Earthwatch Institute, and Dolphin Quest, Inc. P.L.T. was supported by US Office of Naval Research grants N00014-18-1-2062 and N00014-20-1-2709.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Frants Havmand Jensen .

Editor information

Editors and Affiliations

Section Editor information

Rights and permissions

Reprints and permissions

Copyright information

© 2024 Springer Nature Switzerland AG

About this entry

Check for updates. Verify currency and authenticity via CrossMark

Cite this entry

Jensen, F.H. et al. (2024). Automatic Deep-Learning-Based Classification of Bottlenose Dolphin Signature Whistles. In: Popper, A.N., Sisneros, J., Hawkins, A.D., Thomsen, F. (eds) The Effects of Noise on Aquatic Life. Springer, Cham. https://doi.org/10.1007/978-3-031-10417-6_143-1

Download citation

  • DOI: https://doi.org/10.1007/978-3-031-10417-6_143-1

  • Received:

  • Accepted:

  • Published:

  • Publisher Name: Springer, Cham

  • Print ISBN: 978-3-031-10417-6

  • Online ISBN: 978-3-031-10417-6

  • eBook Packages: Springer Reference Biomedicine and Life SciencesReference Module Biomedical and Life Sciences

Publish with us

Policies and ethics