A scalable transfer learning workflow for extracting biological and behavioural insights from forest elephant vocalizations
Open accessAbstract
Animal vocalizations encode rich biological information—such as age, sex, behavioural context and emotional state—making bioacoustic analysis a promising non‐invasive method for assessing welfare and population demography. However, traditional bioacoustic approaches, which rely on manually defined acoustic features, are time‐consuming, require specialized expertise and may introduce subjective bias. These constraints reduce the feasibility of analysing increasingly large datasets generated by passive acoustic monitoring (PAM). Transfer learning with Convolutional Neural Networks (CNNs) offers a scalable alternative by enabling automatic acoustic feature extraction without predefined criteria. Here, we applied four pre‐trained CNNs—two general purpose models (VGGish and YAMNet) and two avian bioacoustic models (Perch and BirdNET)—to African forest elephant ( Loxodonta cyclotis ) recordings. We used a dimensionality reduction algorithm (UMAP) to represent the extracted acoustic features in two dimensions and evaluated these representations across three key tasks: (1) call‐type classification (rumble, roar and trumpet), (2) rumble sub‐type identification and (3) behavioural and demographic analysis. A Random Forest classifier trained on these features achieved near‐perfect accuracy for rumbles, with Perch attaining the highest average accuracy (0.85) across all call types. Clustering the reduced features identified biologically meaningful rumble sub‐types—such as adult female calls linked to logistics—and provided clearer groupings than manual classification. Statistical analyses showed that factors including age and behavioural context significantly influenced call variation ( P < 0.001), with additional comparisons revealing clear differences among contexts (e.g. nursing, competition, separation), sexes and multiple age classes. Perch and BirdNET consistently outperformed general purpose models when dealing with complex or ambiguous calls. These findings demonstrate that transfer learning enables scalable, reproducible bioacoustic workflows capable of detecting biologically meaningful acoustic variation. Integrating this approach into PAM pipelines can enhance the non‐invasive assessment of population dynamics, behaviour and welfare in acoustically active species.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Transfer learning with embeddings from massive avian models like BirdNET or platforms such as OpenSoundscape is surprisingly general: with just a handful of labeled examples per class, researchers can fine-tune these systems for entirely new habitats, species, or whole taxonomic groups.
“Embeddings from massive avian models like BirdNET or platforms such as OpenSoundscape have turned out to be surprisingly general: with just a handful of labeled examples per class, researchers can fine-tune these systems for entirely new habitats, species, or even whole taxonomic groups”
From Listening forward: emerging roles of bioacoustics in ecology, evolution, and conservation (2026) · cites this paper as Pickering et al. 2025
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
Silhouettes: A graphical aid to the interpretation and validation of cluster analysis
Rousseeuw, P. J. (1987) · Journal of Computational and Applied Mathematics · doi:10.1016/0377-0427(87)90125-7 · Cited by 21,392 · see citing works
Principal component analysis: A review and recent developments
Jolliffe, I. T., & Cadima, J. (2016) · Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences · doi:10.1098/rsta.2015.0202 · Cited by 10,117 · see citing works
UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
McInnes, L., Healy, J., & Melville, J. (2018) · arXiv · doi:10.48550/arxiv.1802.03426 · Cited by 7,101 · see citing works
Clustering by Passing Messages Between Data Points
Frey, B. J., & Dueck, D. (2007) · Science · doi:10.1126/science.1136800 · Cited by 6,733 · see citing works
Weiss, K., Khoshgoftaar, T. M., & Wang, D. (2016) · Journal of Big Data · doi:10.1186/s40537-016-0043-6 · Cited by 6,260 · see citing works
Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation
Powers, D. M. W. (2020) · arXiv · doi:10.48550/arxiv.2010.16061 · Cited by 5,440 · see citing works
Dimensionality reduction for visualizing single-cell data using UMAP
Becht, E., McInnes, L., Healy, J., Dutertre, C.-A., Kwok, I. W. H., Ng, L. G., Ginhoux, F., & Newell, E. W. (2019) · Nature Biotechnology · doi:10.1038/nbt.4314 · Cited by 5,358 · see citing works
Audio Set: An ontology and human-labeled dataset for audio events
Gemmeke, J. F., Ellis, D. P. W., Freedman, D., Jansen, A., Lawrence, W., Moore, R. C., Plakal, M., & Ritter, M. (2017) · 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2017.7952261 · Cited by 3,134 · see citing works
librosa: Audio and Music Signal Analysis in Python
McFee, B., Raffel, C., Liang, D., Ellis, D. P. W., McVicar, M., Battenberg, E., & Nieto, O. (2015) · Proceedings of the Python in Science Conferences · doi:10.25080/majora-7b98e3ed-003 · Cited by 3,099 · see citing works
CNN architectures for large-scale audio classification
Hershey, S., Chaudhuri, S., Ellis, D. P. W., Gemmeke, J. F., Jansen, A., Moore, R. C., Plakal, M., Platt, D., Saurous, R. A., Seybold, B., et al. (2017) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2017.7952132 · Cited by 2,605 · see citing works
Bias in error estimation when using cross-validation for model selection
Varma, S., & Simon, R. (2006) · BMC Bioinformatics · doi:10.1186/1471-2105-7-91 · Cited by 2,485 · cited 1× · see citing works
“Nested cross-validation (CV) minimized tuning bias by using inner CV loops for parameter optimization and outer CV loops for error estimation (Varma & Simon, 2006).”
Hutt, S. J. (1977) · Animal Behaviour · doi:10.1016/0003-3472(77)90136-1 · Cited by 2,326 · see citing works
BirdNET: A deep learning solution for avian diversity monitoring
Kahl, S., Wood, C. M., Eibl, M., & Klinck, H. (2021) · Ecological Informatics · doi:10.1016/j.ecoinf.2021.101236 · Cited by 903 · see citing works
Feature dimensionality reduction: a review
Jia, W., Sun, M., Lian, J., & Hou, S. (2022) · Complex & Intelligent Systems · doi:10.1007/s40747-021-00637-x · Cited by 797 · see citing works
Emerging opportunities and challenges for passive acoustics in ecological assessment and monitoring
Gibb, R., Browning, E., Glover‐Kapfer, P., & Jones, K. E. (2018) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13101 · Cited by 753 · see citing works
Self-citationFinding Clusters of Different Sizes, Shapes, and Densities in Noisy, High Dimensional Data
Ertöz, L., Steinbach, M. S., & Kumar, V. (2003) · SIAM International Conference on Data Mining · doi:10.1137/1.9781611972733.5 · Cited by 698 · see citing works
Statistical power for cluster analysis
Dalmaijer, E. S., Nord, C. L., & Astle, D. E. (2022) · BMC Bioinformatics · doi:10.1186/s12859-022-04675-1 · Cited by 600 · see citing works
Tunability: Importance of Hyperparameters of Machine Learning AlgorithmsPreprint
Probst, P., Bischl, B., & Boulesteix, A. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1802.09596 · Cited by 590 · see citing works
Overview and comparative study of dimensionality reduction techniques for high dimensional data
Ayesha, S., Hanif, M. K., & Talib, R. (2020) · Information Fusion · doi:10.1016/j.inffus.2020.01.005 · Cited by 579 · see citing works
Computational bioacoustics with deep learning: a review and roadmap
Stowell, D. (2022) · PeerJ · doi:10.7717/peerj.13152 · Cited by 471 · see citing works
Finding, visualizing, and quantifying latent structure across diverse animal vocal repertoires
Sainburg, T., Thielk, M., & Gentner, T. Q. (2020) · PLoS Computational Biology · doi:10.1371/journal.pcbi.1008228 · Cited by 324 · see citing works
Pitfalls in the categorization of behaviour: a comparison of dolphin whistle classification methods
Janik, V. M. (1999) · Animal Behaviour · doi:10.1006/anbe.1998.0923 · Cited by 198 · see citing works
Acoustic monitoring for conservation in tropical forests: examples from forest elephants
Wrege, P. H., Rowland, E. D., Keen, S., & Shiu, Y. (2017) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.12730 · Cited by 191 · see citing works
Characterizing soundscapes across diverse ecosystems using a universal acoustic feature set
Sethi, S. S., Jones, N. S., Fulcher, B., Picinali, L., Clink, D. J., Klinck, H., Orme, C. D. L., Wrege, P. H., & Ewers, R. M. (2020) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.2004702117 · Cited by 171 · see citing works
Global birdsong embeddings enable superior transfer learning for bioacoustic classification
Ghani, B., Denton, T., Kahl, S., & Klinck, H. (2023) · Scientific Reports · doi:10.1038/s41598-023-49989-z · Cited by 167 · see citing works
Universal mechanisms of sound production and control in birds and mammals
Elemans, C. P. H., Rasmussen, J. H., Herbst, C. T., Düring, D. N., Zollinger, S. A., Brumm, H., Srivastava, K. H., Svane, N., Ding, M. L., Larsen, O. N., et al. (2015) · Nature Communications · doi:10.1038/ncomms9978 · Cited by 163 · see citing works
Mcloughlin, M. P., Stewart, R., & McElligott, A. G. (2019) · Journal of The Royal Society Interface · doi:10.1098/rsif.2019.0225 · Cited by 156 · see citing works
ORCA-SPOT: An Automatic Killer Whale Sound Detection Toolkit Using Deep Learning
Bergler, C., Schröter, H., Cheng, R. X., Barth, V., Weber, M., Nöth, E., Hofer, H., & Maier, A. (2019) · Scientific Reports · doi:10.1038/s41598-019-47335-w · Cited by 127 · see citing works
Goffinet, J., Brudner, S. N., MOONEY, R. D., & Pearson, J. (2021) · eLife · doi:10.7554/elife.67855 · Cited by 111 · see citing works
Briefer, E. F., Sypherd, C. C.-R., Linhart, P., Leliveld, L. M. C., de la Torre, M. P., Read, E. R., Guérin, C., Deiss, V., Monestier, C., Rasmussen, J. H., et al. (2022) · Scientific Reports · doi:10.1038/s41598-022-07174-8 · Cited by 107 · see citing works
Passive acoustic monitoring of animal populations with transfer learning
Dufourq, E., Batist, C., Foquet, R., & Durbach, I. (2022) · Ecological Informatics · doi:10.1016/j.ecoinf.2022.101688 · Cited by 97 · see citing works
Western Gorilla Vocal Repertoire and Contextual Use of Vocalizations
Salmi, R., Hammerschmidt, K., & Doran‐Sheehy, D. M. (2013) · Ethology · doi:10.1111/eth.12122 · Cited by 89 · see citing works
Temporal stability and change in the social call repertoire of migrating humpback whales
Rekdahl, M. L., Dunlop, R. A., Noad, M. J., & Goldizen, A. W. (2013) · The Journal of the Acoustical Society of America · doi:10.1121/1.4789941 · Cited by 88 · see citing works
Internal states and extrinsic factors both determine monkey vocal production
Liao, D. A., Zhang, Y. S., Cai, L. X., & Ghazanfar, A. A. (2018) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1722426115 · Cited by 88 · see citing works
How assessing relationships between emotions and cognition can improve farm animal welfare
Boissy, A., & Lee, C. (2014) · Scientific and Technical Review (World Organisation for Animal Health) · doi:10.20506/rst.33.1.2260 · Cited by 80 · see citing works
Visualizing Sound Emission of Elephant Vocalizations: Evidence for Two Rumble Production Types
Stoeger, A. S., Heilmann, G., Zeppelzauer, M., Ganswindt, A., Hensman, S., & Charlton, B. D. (2012) · PLoS ONE · doi:10.1371/journal.pone.0048907 · Cited by 79 · see citing works
Leong, K. M., Ortolani, A., Burks, K. D., Mellen, J. D., & Savage, A. (2003) · Bioacoustics · doi:10.1080/09524622.2003.9753499 · Cited by 77 · see citing works
Acoustic communication in the Asian elephant, Elephas maximus maximus
de Silva, S. (2010) · Behaviour · doi:10.1163/000579510x495762 · Cited by 71 · see citing works
Behavioral Contexts of Elephant Acoustic Communication
Poole, J. H. (2011) · University of Chicago Press eBooks · doi:10.7208/chicago/9780226542263.003.0009 · Cited by 69 · see citing works
Nair, S., Balakrishnan, R., Seelamantula, C. S., & Sukumar, R. (2009) · The Journal of the Acoustical Society of America · doi:10.1121/1.3224717 · Cited by 66 · see citing works
Soltis, J. M., Blowers, T. E., & Savage, A. (2011) · The Journal of the Acoustical Society of America · doi:10.1121/1.3531798 · Cited by 61 · see citing works
Wood, C. M., Klinck, H., Gustafson, M. L., Keane, J. J., Sawyer, S. C., Gutiérrez, R. J., & Peery, M. Z. (2020) · Conservation Biology · doi:10.1111/cobi.13516 · Cited by 59 · see citing works
Exploring Animal Behavior Through Sound: Volume 1
Erbe, C., & Thomas, J. A. (2022) · Springer eBooks · doi:10.1007/978-3-030-97540-1 · Cited by 59 · see citing works
Napier, T., Ahn, E., Allen‐Ankins, S., Schwarzkopf, L., & Lee, I. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.124220 · Cited by 55 · see citing works
Stoeger, A. S., Zeppelzauer, M., & Baotic, A. (2014) · Bioacoustics · doi:10.1080/09524622.2014.888375 · Cited by 51 · see citing works
WOOD, J. D., McCowan, B., Langbauer, W. R., Viljoen, J. J., & Hart, L. A. (2005) · Bioacoustics · doi:10.1080/09524622.2005.9753544 · Cited by 49 · see citing works
Elephants and algorithms: a review of the current and future role of AI in elephant monitoring
Brickson, L. L., Zhang, L., Vollrath, F. W., Douglas‐Hamilton, I., & Titus, A. (2023) · Journal of The Royal Society Interface · doi:10.1098/rsif.2023.0367 · Cited by 44 · see citing works
Assessing inter-annotator agreement from collaborative annotation campaign in marine bioacoustics
Duc, P. N. H., Torterotot, M., Samaran, F., White, P. R., Gérard, O., Adam, O., & Cazau, D. (2020) · Ecological Informatics · doi:10.1016/j.ecoinf.2020.101185 · Cited by 40 · see citing works
Differences in combinatorial calls among the 3 elephant species cannot be explained by phylogeny
Pardo, M. A., Poole, J. H., Stoeger, A. S., Wrege, P. H., O'Connell‐Rodwell, C. E., Padmalal, U. K. G. K., & de Silva, S. (2019) · Behavioral Ecology · doi:10.1093/beheco/arz018 · Cited by 39 · see citing works
McCordic, J., Root‐Gutteridge, H., Cusano, D. A., Denes, S., & Parks, S. E. (2016) · Endangered Species Research · doi:10.3354/esr00735 · Cited by 32 · see citing works
Pal, K., & Sharma, M. (2020) · International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC) · doi:10.1109/i-smac49090.2020.9243502 · Cited by 23 · see citing works
African and Asian Elephant Vocal Communication: A Cross-Species Comparison
Stoeger, A. S., & de Silva, S. (2013) · Springer eBooks · doi:10.1007/978-94-007-7414-8_3 · Cited by 22 · see citing works
Vocal expression of emotional arousal across two call types in young rhesus macaques
Schwartz, J. W., Sánchez, M. M., & Gouzoules, H. (2022) · Animal Behaviour · doi:10.1016/j.anbehav.2022.05.017 · Cited by 22 · see citing works
Acoustic structure of forest elephant rumbles: a test of the ambiguity reduction hypothesis
Hedwig, D., Verahrami, A. K., & Wrege, P. H. (2019) · Animal Cognition · doi:10.1007/s10071-019-01304-y · Cited by 20 · see citing works
Self-citationDoes Social Complexity Drive Vocal Complexity? Insights from the Two African Elephant Species
Hedwig, D., Poole, J. H., & Granli, P. (2021) · Animals · doi:10.3390/ani11113071 · Cited by 20 · see citing works
Self-citationBinary Glioma Grading: Radiomics versus Pre-trained CNN Features
Decuyper, M., Bonte, S., & Holen, R. V. (2018) · Lecture notes in computer science · doi:10.1007/978-3-030-00931-1_57 · Cited by 18 · see citing works
Scalable preprocessing of high volume environmental acoustic data for bioacoustic monitoring
Brown, A. F., Garg, S., & Montgomery, J. (2018) · PLoS ONE · doi:10.1371/journal.pone.0201542 · Cited by 17 · see citing works
Call type repertoire of killer whales (Orcinus orca) in Iceland and its variation across regions
Selbmann, A., Deecke, V. B., Filatova, O. A., Дмитриевич, Ф. И., Miller, P. J. O., Simon, M. J., Bowles, A. E., Lyrholm, T., Lacey, C., Magnúsdóttir, E., et al. (2023) · Marine Mammal Science · doi:10.1111/mms.13039 · Cited by 14 · see citing works
Call combination in African forest elephants Loxodonta cyclotis
Hedwig, D., & Kohlberg, A. B. (2024) · PLoS ONE · doi:10.1371/journal.pone.0299656 · Cited by 12 · see citing works
Self-citationVocal Creativity in Elephant Sound Production
Stoeger, A. S., Baotic, A., & Heilmann, G. (2021) · Biology · doi:10.3390/biology10080750 · Cited by 11 · see citing works
Whombat: An open‐source audio annotation tool for machine learning assisted bioacoustics
Martínez Balvanera, S., Mac Aodha, O., Weldy, M. J., Pringle, H., Browning, E., & Jones, K. E. (2025) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.14468 · Cited by 11 · see citing works
Self-citationElephant sonic and infrasonic sound production, perception, and processing
Stoeger, A. S. (2021) · Elsevier eBooks · doi:10.1016/b978-0-12-815160-0.00023-2 · Cited by 10 · see citing works
Automatic Detection and Compression for Passive Acoustic Monitoring of the African Forest Elephant
Björck, J., Rappazzo, B., Chen, D., Bernstein, R. J., Wrege, P. H., & Gomes, C. P. (2019) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v33i01.3301476 · Cited by 8 · see citing works
Lokhandwala, S., Sinha, R., Sreeram, G., & Pailla, B. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-48312-7_30 · Cited by 5 · see citing works
Hasan, N. I. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2201.00124 · Cited by 4 · see citing works
Flagged first, then by citations
Research works printed but not linked: 3
- 000579510X495762 · DOI
- Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N. et al. (2017) Attention is all you need. Advances in Neural Information Processing Systems [Preprint].
- Yosinski, J., Clune, J., Bengio, Y. & Lipson, H. (2014) How transferable are features in deep neural networks? In: Advances in neural information processing systems. Montreal, Canada: Curran Associates, Inc. Available at: https:// proceedings.neurips.cc/paper_files/paper/2014/hash/ 375c71349b295fbe2dcdca9206f20a06-Abstract.html [Accessed 5 January 2025]
Cited by
Sorted by citations
The sound of silence: Sex and individual differences in rat ultrasonic vocalisation
Quante, S. M., Groneberg, E., Kaiser, S., & Richter, S. H. (2025) · Behavioural Brain Research · doi:10.1016/j.bbr.2025.115823 · Cited by 4 · see citing works
Automated bird sound recognition in ecology: A methodological review and application framework
Shen, X., Xu, J., Zhang, Q., Luo, P., Liu, Z., Chen, J., Zhou, G., Yi, J., Xiao, Z., & Chen, A. (2026) · Avian Research · doi:10.1016/j.avrs.2026.100368 · Cited by 4 · see citing works
Syunkova, A., Lapp, S., Basanta, M. D., Lambertini, C., Guzman, S. R., Voyles, J., Richards‐Zawacki, C. L., & Kitzes, J. A. (2025) · Ecological Informatics · doi:10.1016/j.ecoinf.2025.103427 · Cited by 1 · see citing works
Listening forward: emerging roles of bioacoustics in ecology, evolution, and conservation
Zsebők, S. (2026) · Biologia Futura · doi:10.1007/s42977-026-00322-5
“…or platforms such as OpenSoundscape have turned out to be surprisingly general: with just a handful of labeled examples per class, researchers can fine-tune these systems for entirely new habitats, species, or even whole taxonomic groups (Ghani et al. 2023; Lapp et al. 2023; Pickering et al. 2025).”
Kanna, R. D., & Aruna, R. (2025) · IEEE International Conference on Information Technology, Electronics and Intelligent Communication Systems (ICITEICS) · doi:10.1109/iciteics64870.2025.11341052
- Authors
- Pickering, A., Martinez Balvanera, S., Jones, K. E., & Hedwig, D.
- Affiliations
- University College London (GB)
- Centre for Biodiversity and Environment Research, Department of Genetics, Evolution and Environment University College London London WC1E 6BT UK
- Cornell University (US)
- Corresponding author
- Pickering, Alastair
- Venue
- Remote Sensing in Ecology and Conservation
- Publisher
- Wiley
- Published
- 2025-04-25
- Volume
- 11
- Issue
- 5
- Pages
- 590-605
- ISSN
- 2056-3485 (ISSN-L)
- OpenAlex
- W4409795067
- Cited by
- 5
- References
- 64
- Citations by year
- 2026: 2 · 2025: 3
- FWCI
- 4.56
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Remote Sensing in Ecology and Conservation (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- UCL Discovery (University College London) (submittedVersion)
- Funders
- National Geographic Society (28718)
- Natural Environment Research Council (NE/S007229/1)
- Consejo Nacional de Humanidades, Ciencias y Tecnologías (2020‐000017‐02EXTF‐00334)
- Topics
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Developmental Biology › Animal Vocal Communication and Behavior
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Physical Sciences › Environmental Science › Ecology › Marine animal studies overview
- Keywords
- African forest elephant, Loxodonta cyclotis, animal vocalizations, transfer learning, convolutional neural networks, behavioral context, animal welfare, demography, passive acoustic monitoring, perch, BirdNET, VGGish, YAMNet
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
