Many animals emit vocal sounds which, independently from the sounds' function, contain some individually distinctive signature. Thus the automatic recognition of individuals by sound is a potentially powerful tool for zoology and ecology research and practical monitoring. Here, we present a general automatic identification method that can work across multiple animal species with various levels of complexity in their communication systems. We further introduce new analysis techniques based on dataset manipulations that can evaluate the robustness and generality of a classifier. By using these techniques, we confirmed the presence of experimental confounds in situations resembling those from past studies. We introduce data manipulations that can reduce the impact of these confounds, compatible with any classifier. We suggest that assessment of confounds should become a standard part of future studies to ensure they do not report over-optimistic results. We provide annotated recordings used for analyses along with this study and we call for dataset sharing to be a common practice to enhance the development of methods and comparisons of results.

Department of Behavioural Ecology Faculty of Biology Adam Mickiewicz University Poznań Poland

Department of Ecology Faculty of Science Charles University Prague Czech Republic

Faculty of Environmental Sciences Czech University of Life Sciences Prague Prague Czech Republic

Institute of Vertebrate Biology The Czech Academy of Sciences Brno Czech Republic

Machine Listening Lab Centre for Digital Music Queen Mary University of London London UK

See more in PubMed

Amorim MCP, Vasconcelos RO. 2008. Variability in the mating calls of the Lusitanian toadfish Halobatrachus didactylus: cues for potential individual recognition. J. Fish Biol. 73, 1267–1283. (10.1111/jfb.2008.73.issue-6) DOI

Bee MA, Gerhardt HC. 2001. Neighbour-stranger discrimination by territorial male bullfrogs (Rana catesbeiana): I. Acoustic basis. Anim. Behav. 62, 1129–1140. (10.1006/anbe.2001.1851) DOI

Terry AM, Peake TM, McGregor PK. 2005. The role of vocal individuality in conservation. Front. Zool. 2, 10 (10.1186/1742-9994-2-10) PubMed DOI PMC

Taylor AM, Reby D. 2010. The contribution of source-filter theory to mammal vocal communication research. J. Zool. 280, 221–236. (10.1111/j.1469-7998.2009.00661.x) DOI

Gamba M, Favaro L, Araldi A, Matteucci V, Giacoma C, Friard O. 2017. Modeling individual vocal differences in group-living lemurs using vocal tract morphology. Curr. Zool. 63, 467–475. (10.1093/cz/zox023) PubMed DOI PMC

Janik V, Slater PB. 1997. Vocal learning in mammals, vol. 26, pp. 59–99. New York, NY: Academic Press.

Wiley RH. 2013. Specificity and multiplicity in the recognition of individuals: implications for the evolution of social behaviour. Biol. Rev. 88, 179–195. (10.1111/brv.2013.88.issue-1) PubMed DOI

Boeckle M, Bugnyar T. 2012. Long-term memory for affiliates in ravens. Curr. Biol. 22, 801–806. (10.1016/j.cub.2012.03.023) PubMed DOI PMC

Insley SJ. 2000. Long-term vocal recognition in the northern fur seal. Nature 406, 404–405. (10.1038/35019064) PubMed DOI

Briefer EF, de la Torre MP, McElligott AG. 2012. Mother goats do not forget their kids’ calls. Proc. R. Soc. B 279, 3749–3755. (10.1098/rspb.2012.0986) PubMed DOI PMC

Slabbekoorn H. 2004. Singing in the wild: the ecology of birdsong. In Nature’s music: the science of birdsong (eds P Marler, H Slabbekoorn), pp. 178–205. San Diego, CA: Elsevier Academic Press.

Mouterde SC, Elie JE, Theunissen FE, Mathevon N. 2014. Learning to cope with degraded sounds: female zebra finches can improve their expertise at discriminating between male voices at long distance. J. Exp. Biol. 217, 3169–3177. (10.1242/jeb.104463) PubMed DOI PMC

Gambale PG, Signorelli L, Bastos RP. 2014. Individual variation in the advertisement calls of a Neotropical treefrog (Scinax constrictus). Amphibia-Reptilia 35, 271–281. (10.1163/15685381-00002949) DOI

Collins SA. 2004. Vocal fighting and flirting: the functions of birdsong. In Nature’s music: the science of birdsong (eds PR Marler, H Slabbekoorn), pp. 39–79. Elsevier Academic Press.

Linhart P, Jaška P, Petrusková T, Petrusek A, Fuchs R. 2013. Being angry, singing fast? Signalling of aggressive motivation by syllable rate in a songbird with slow song. Behav. Processes 100, 139–145. (10.1016/j.beproc.2013.06.012) PubMed DOI

Kroodsma DE. 2004. The diversity and plasticity of bird song. In Nature’s music: the science of birdsong (eds PR Marler, H Slabbekoorn), pp. 108–131. Elsevier Academic Press.

Thom MDF, Dytham C. 2012. Female choosiness leads to the evolution of individually distinctive males. Evolution 66, 3736–3742. (10.1111/evo.2012.66.issue-12) PubMed DOI

Bradbury JW, Vehrencamp SL. 1998. Principles of animal communication, 1st edn Sunderland, MA: Sinauer Associates.

Crowley PH, Provencher L, Sloane S, Dugatkin LA, Spohn B, Rogers L, Alfieri M. 1996. Evolving cooperation: the role of individual recognition. Biosystems 37, 49–66. (10.1016/0303-2647(95)01546-9) PubMed DOI

Mennill DJ. 2011. Individual distinctiveness in avian vocalizations and the spatial monitoring of behaviour. Ibis 153, 235–238. (10.1111/j.1474-919X.2011.01119.x) DOI

Blumstein DT. et al. 2011. Acoustic monitoring in terrestrial environments using microphone arrays: applications, technological considerations and prospectus. J. Appl. Ecol. 48, 758–767. (10.1111/j.1365-2664.2011.01993.x) DOI

Johnsen A, Lifjeld J, Rohde PA. 1997. Coloured leg bands affect male mate-guarding behaviour in the bluethroat. Anim. Behav. 54, 121–130. (10.1006/anbe.1996.0437) PubMed DOI

Gervais JA, Catlin DH, Chelgren ND, Rosenberg DK. 2006. Radiotransmitter mount type affects burrowing owl survival. J. Wildl. Manag. 70, 872–876. (10.2193/0022-541X(2006)70[872:RMTABO]2.0.CO;2) DOI

Linhart P, Fuchs R, Poláková S, Slabbekoorn H. 2012. Once bitten twice shy: long-term behavioural changes caused by trapping experience in willow warblers Phylloscopus trochilus. J. Avian Biol. 43, 186–192. (10.1111/j.1600-048X.2012.05580.x) DOI

Rivera-Gutierrez HF, Pinxten R, Eens M. 2015. Songbirds never forget: long-lasting behavioural change triggered by a single playback event. Behaviour 152, 1277–1290. (10.1163/1568539X-00003278) DOI

Camacho C, Canal D, Potti J. 2017. Lifelong effects of trapping experience lead to age-biased sampling: lessons from a wild bird population. Anim. Behav. 130, 133–139. (10.1016/j.anbehav.2017.06.018) DOI

Petrusková T, Pišvejcová I, Kinštová A, Brinke T, Petrusek A. 2015. Repertoire-based individual acoustic monitoring of a migratory passerine bird with complex song as an efficient tool for tracking territorial dynamics and annual return rates. Methods Ecol. Evol. 7, 274–284. (10.1111/2041-210X.12496) DOI

Laiolo P, Vögeli M, Serrano D, Tella JL. 2007. Testing acoustic versus physical marking: two complementary methods for individual-based monitoring of elusive species. J. Avian Biol. 38, 672–681. (10.1111/jav.2007.38.issue-6) DOI

Kirschel ANG, Cody ML, Harlow ZT, Promponas VJ, Vallejo EE, Taylor CE. 2011. Territorial dynamics of Mexican Ant-thrushes Formicarius moniliger revealed by individual recognition of their songs. Ibis 153, 255–268. (10.1111/j.1474-919X.2011.01102.x) DOI

Spillmann B, van Schaik CP, Setia TM, Sadjadi SO. 2017. Who shall I say is calling? Validation of a caller recognition procedure in Bornean flanged male orangutan (Pongo pygmaeus wurmbii) long calls. Bioacoustics 26, 109–120. (10.1080/09524622.2016.1216802) DOI

Delport W, Kemp AC, Ferguson JWH. 2002. Vocal identification of individual African Wood Owls Strix woodfordii: a technique to monitor long-term adult turnover and residency. Ibis 144, 30–39. (10.1046/j.0019-1019.2001.00019.x) DOI

Adi K, Johnson MT, Osiejuk TS. 2010. Acoustic censusing using automatic vocalization classification and identity recognition. J. Acoust. Soc. Am. 127, 874–883. (10.1121/1.3273887) PubMed DOI

Terry AMR, McGregor PK. 2002. Census and monitoring based on individually identifiable vocalizations: the role of neural networks. Anim. Conserv. 5, 103–111. (10.1017/S1367943002002147) DOI

Fox EJS. 2008. A new perspective on acoustic individual recognition in animals with limited call sharing or changing repertoires. Anim. Behav. 75, 1187–1194. (10.1016/j.anbehav.2007.11.003) DOI

Fox EJS, Roberts JD, Bennamoun M. 2008. Call-independent individual identification in birds. Bioacoustics 18, 51–67. (10.1080/09524622.2008.9753590) DOI

Cheng J, Sun Y, Ji L. 2010. A call-independent and automatic acoustic system for the individual recognition of animals: a novel model using four passerines. Pattern Recognit. 43, 3846–3852. (10.1016/j.patcog.2010.04.026) DOI

Cheng J, Xie B, Lin C, Ji L. 2012. A comparative study in birds: call-type-independent species and individual recognition using four machine-learning methods and two acoustic features. Bioacoustics 21, 157–171. (10.1080/09524622.2012.669664) DOI

Szegedy C, Zaremba W, Sutskever I, Bruna J, Erhan D, Goodfellow I, Fergus R. 2013. Intriguing properties of neural networks. (http://arxiv.org/abs/13126199).

Mesaros A, Heittola T, Virtanen T. 2018. Acoustic Scene Classification: an Overview of DCASE 2017 Challenge Entries. In 16th Int. Workshop on Acoustic Signal Enhancement (IWAENC). Tokyo, Japan.

Khanna H, Gaunt S, McCallum D. 1997. Digital spectrographic cross-correlation: tests of sensitivity. Bioacoustics 7, 209–234. (10.1080/09524622.1997.9753332) DOI

Foote JR, Palazzi E, Mennill DJ. 2013. Songs of the Eastern Phoebe, a suboscine songbird, are individually distinctive but do not vary geographically. Bioacoustics 22, 137–151. (10.1080/09524622.2012.740174) DOI

Ptáček L, Machlica L, Linhart P, Jaška P, Muller L. 2016. Automatic recognition of bird individuals on an open set using as-is recordings. Bioacoustics 25, 55–73. (10.1080/09524622.2015.1089524) DOI

Stowell D, Stylianou Y, Wood M, Pamuła H, Glotin H. 2019. Automatic acoustic detection of birds through deep learning: the first Bird Audio Detection challenge. Methods Ecol. Evol. 10, 363–380. (10.1111/2041-210X.13103) DOI

Knight EC, Hannah KC, Foley GJ, Scott CD, Brigham RM, Bayne E. 2017. Recommendations for acoustic recognizer performance assessment with application to five common automated signal recognition programs. Avian Conserv. Ecol. 12, 14 (10.5751/ACE-01114-120214) DOI

Joly A, Goëau H, Botella C, Glotin H, Bonnet P, Vellinga WP. et al. 2018. Overview of LifeCLEF 2018: a large-scale evaluation of species identification and recommendation algorithms in the Era of AI. In Int. Conf. of the Cross-Language Evaluation Forum for European Languages, pp. 247–266. Berlin, Germany: Springer.

Stowell D. 2018. Computational bioacoustic scene analysis. In Computational analysis of sound scenes and events, pp. 303–333. Berlin, Germany: Springer.

Morfi V, Stowell D. 2018. Deep learning for audio event detection and tagging on low-resource datasets. Appl. Sci. 8, 1397 (10.3390/app8081397) DOI

Salamon J, Bello JP. 2017. Deep convolutional neural networks and data augmentation for environmental sound classification. IEEE Signal Process Lett. 24, 279–283. (10.1109/LSP.2017.2657381) DOI

Lasseck M. 2018. Audio-based bird species identification with deep convolutional neural networks. Working Notes of CLEF. 2018.

Grava T, Mathevon N, Place E, Balluet P. 2008. Individual acoustic monitoring of the European Eagle Owl Bubo bubo. Ibis 150, 279–287. (10.1111/j.1474-919X.2007.00776.x) DOI

Petrusková T, Osiejuk TS, Linhart P, Petrusek A. 2008. Structure and complexity of perched and flight songs of the tree pipit (Anthus trivialis). Ann. Zool. Fennici 45, 135–148. (10.5735/086.045.0205) DOI

Průchová A, Jaška P, Linhart P. 2017. Cues to individual identity in songs of songbirds: testing general song characteristics in Chiffchaffs Phylloscopus collybita. J. Ornithol. 158, 911–924. (10.1007/s10336-017-1455-6) DOI

Nieuwenhuyse DV, Génot JC, Johnson DH. 2008. The little owl: conservation, ecology and behavior of Athene noctua. Cambridge, UK: Cambridge University Press.

Linhart P, Šálek M. 2017. The assessment of biases in the acoustic discrimination of individuals. PLoS ONE 12, e0177206 (10.1371/journal.pone.0177206) PubMed DOI PMC

Krizhevsky A, Sutskever I, Hinton GE. 2012. ImageNet classification with deep convolutional neural networks. In Advances in neural information processing systems (NIPS), pp. 1097–1105. See http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.

Cireşan D, Meier U, Schmidhuber J. Multi-column deep neural networks for image classification. (http://arxiv.org/abs/12022745). PubMed

Schlüter J, Grill T. 2015. Exploring data augmentation for improved singing voice detection with neural networks. In Proc. of the Int. Conf. on Music Information Retrieval (ISMIR), pp. 121–126.

Liang D, Yang F, Zhang T, Yang P. 2018. Understanding mixup training methods. IEEE Access 6, 58 774–58 783. (10.1109/ACCESS.2018.2872698) DOI

Stowell D, Plumbley MD. 2014. Automatic large-scale classification of bird sounds is strongly improved by unsupervised feature learning. PeerJ 2, e488 (10.7717/peerj.488) PubMed DOI PMC

Fawcett T. 2006. An introduction to ROC analysis. Pattern Recognit. Lett. 27, 861–874. (10.1016/j.patrec.2005.10.010) DOI

Fournier DA, Skaug HJ, Ancheta J, Ianelli J, Magnusson A, Maunder MN, Nielsen A, Sibert J. 2012. AD model builder: using automatic differentiation for statistical inference of highly parameterized complex nonlinear models. Optim. Methods Softw. 27, 233–249. (10.1080/10556788.2011.597854) DOI

Skaug H, Fournier D, Bolker B, Magnusson A, Nielsen A. Generalized linear mixed models using ‘AD Model Builder’; 2016-01-19. R package version 0.8.3.3.

Beecher MD, Brenowitz EA. 2005. Functional aspects of song learning in songbirds. Trends Ecol. Evol. 20, 143–149. (10.1016/j.tree.2005.01.004) PubMed DOI