A Convolutional Neural Network Bird Species Recognizer Built From Little Data by Iteratively Training, Detecting, and Labeling
Open accessAbstract
Automatically detecting the calls of species of interest in audio recordings is a common but often challenging exercise in ecoacoustics. This challenge is increasingly being tackled with deep neural networks that generally require a rich set of training data. Often, the available training data might not be from the same geographical region as the study area and so may contain important differences. This mismatch in training and deployment datasets can impact the accuracy at deployment, mainly due to confusing sounds absent from the training data generating false positives, as well as some variation in call types. We have developed a multiclass convolutional neural network classifier for seven target bird species to track presence absence of these species over time in cotton growing regions. We started with no training data from cotton regions but we did have an unbalanced library of calls from other locations. Due to the relative scarcity of calls in recordings from cotton regions, manually scanning and labeling the recordings was prohibitively time consuming. In this paper we describe our process of overcoming this data mismatch to develop a recognizer that performs well on the cotton recordings for most classes. The recognizer was trained on recordings from outside the cotton regions and then applied to unlabeled cotton recordings. Based on the resulting outputs a verification set was chosen to be manually tagged and incorporated in the training set. By iterating this process, we were gradually able to build the training set of cotton audio examples. Through this process, we were able to increase the average class F1 score (the harmonic mean of precision and recall) of the recognizer on target recordings from 0.45 in the first iteration to 0.74.
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?
Machine learning techniques are revolutionising acoustic data processing through the development of automatic detection, measurement, and species-specific call recognisers.
“However, machine learning techniques to process acoustic data are revolutionising such studies, through the development of techniques including automatic detection, measurement, and the application of species‐specific call recognisers (Bianco et al. 2019; Teixeira et al. 2019; Eichinski et al. 2022; Stowell 2022).”
From Roars, Rumbles, and Resonance: A Systematic Review and Meta‐Analysis of Crocodylian Acoustic Signals (2026) · cites this paper as Eichinski et al. 2022
The Abstract describes a neural-network recognizer that automatically detects calls from seven bird species, and Related Research describes advances in automated acoustic analysis. These passages support detection and species-specific call recognition, but they do not establish the claim about automatic measurement; the paper also does not itself characterize these techniques as revolutionising acoustic studies. (LLM, conf=0.84)
Experimental — accuracy not yet measuredMachine learning for call detection and identification has traditionally required a significant amount of pre-processed data.
“Machine learning for call detection and identification has traditionally required a significant amount of pre‐processed data”
From Roars, Rumbles, and Resonance: A Systematic Review and Meta‐Analysis of Crocodylian Acoustic Signals (2026) · cites this paper as Eichinski et al. 2022
Recent breakthroughs have demonstrated that less data can be sufficient for machine learning call detection and identification, depending on research goals.
“but recent breakthroughs have demonstrated that less data can be sufficient, depending on research goals (Knight et al. 2020; Eichinski et al. 2022).”
From Roars, Rumbles, and Resonance: A Systematic Review and Meta‐Analysis of Crocodylian Acoustic Signals (2026) · cites this paper as Eichinski et al. 2022
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
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · see citing works
Improving Generalization with Active Learning
Cohn, D. A., Atlas, L., & Ladner, R. E. (1994) · Machine Learning · doi:10.1023/a:1022673506211 · Cited by 1,597 · see citing works
Fastai: A Layered API for Deep Learning
Howard, J., & Gugger, S. (2020) · Information · doi:10.3390/info11020108 · Cited by 992 · see citing works
A Review of Domain Adaptation without Target Labels
Kouw, W. M., & Loog, M. (2021) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2019.2945942 · Cited by 613 · see citing works
Both, C., van Turnhout, C. A. M., Bijlsma, R. G., Siepel, H., Strien, A. J. V., & Foppen, R. (2009) · Proceedings of the Royal Society B Biological Sciences · doi:10.1098/rspb.2009.1525 · Cited by 602 · see citing works
Kumar, A., V, Prakash, G., & Kumar, G. (2020) · Journal of Retailing and Consumer Services · doi:10.1016/j.jretconser.2020.102270 · Cited by 421 · see citing works
Stowell, D., Wood, M. D., Pamuła, H., Stylianou, Y., & Glotin, H. (2019) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13103 · Cited by 414 · see citing works
Measuring Domain Shift for Deep Learning in Histopathology
Stacke, K., Eilertsen, G., Unger, J., & Lundström, C. F. (2020) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.3032060 · Cited by 334 · see citing works
Acevedo, M. A., & Villanueva-Rivera, L. J. (2006) · Wildlife Society Bulletin · doi:10.2193/0091-7648(2006)34[211:uadrsa]2.0.co;2 · Cited by 231 · see citing works
Investigation of Different CNN-Based Models for Improved Bird Sound Classification
Xie, J., Hu, K., Zhu, M., Yu, J., & Zhu, Q. (2019) · IEEE Access · doi:10.1109/access.2019.2957572 · Cited by 145 · see citing works
Comparing recurrent convolutional neural networks for large scale bird species classification
Gupta, G., Kshirsagar, M., Zhong, M., Gholami, S., & Ferres, J. L. (2021) · Scientific Reports · doi:10.1038/s41598-021-96446-w · Cited by 132 · see citing works
The Australian Acoustic Observatory
Roe, P., Eichinski, P., Fuller, R. A., McDonald, P. G., Schwarzkopf, L., Towsey, M., Truskinger, A., Tucker, D., & Watson, D. M. (2021) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13660 · Cited by 91 · see citing works
Self-citationPreventing dataset shift from breaking machine-learning biomarkers
Dockès, J., Varoquaux, G. P., & Poline, J. (2021) · GigaScience · doi:10.1093/gigascience/giab055 · Cited by 86 · see citing works
García, K., Olimpi, E. M., Karp, D. S., & Gonthier, D. J. (2020) · Journal of Integrated Pest Management · doi:10.1093/jipm/pmaa009 · Cited by 81 · see citing works
Analysing environmental acoustic data through collaboration and automation
Wimmer, J., Towsey, M., Planitz, B. M., Williamson, I., & Roe, P. (2012) · Future Generation Computer Systems · doi:10.1016/j.future.2012.03.004 · Cited by 69 · see citing works
Self-citationActive learning for bird sound classification via a kernel-based extreme learning machine
Qian, K., Zhang, Z., Baird, A., & Schuller, B. W. (2017) · The Journal of the Acoustical Society of America · doi:10.1121/1.5004570 · Cited by 56 · see citing works
Simultaneous segmentation and classification of bird song using CNN
Narasimhan, R., Fern, X. Z., & Raich, R. (2017) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2017.7952135 · Cited by 54 · see citing works
A multi-view CNN-based acoustic classification system for automatic animal species identification
Xu, W., Zhang, X., Yao, L., Xue, W., & Wei, B. W. (2020) · Ad Hoc Networks · doi:10.1016/j.adhoc.2020.102115 · Cited by 44 · see citing works
Newell, F. L., Sheehan, J., Wood, P. B., Rodewald, A. D., Buehler, D. A., Keyser, P. D., Larkin, J. L., Beachy, T. A., Bakermans, M. H., Boves, T. J., et al. (2013) · Journal of Field Ornithology · doi:10.1111/jofo.12026 · Cited by 35 · see citing works
Wang, Y., Méndez, A. E. M., Cartwright, M., & Bello, J. P. (2019) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2019.8683063 · Cited by 27 · see citing works
BioSounds: an open-source, online platform for ecoacoustics
Darras, K., Pérez, N., Mauladi, & Hanf‐Dressler, T. (2020) · F1000Research · doi:10.12688/f1000research.26369.1 · Cited by 24 · see citing works
Active learning for classifying long‐duration audio recordings of the environment
Kholghi, M., Phillips, Y., Towsey, M., Sitbon, L., & Roe, P. (2018) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13042 · Cited by 22 · see citing works
Self-citationLeseberg, N. P., Venables, W. N., Murphy, S., & Watson, J. E. M. (2020) · Methods in Ecology and Evolution · doi:10.1111/2041-210x.13475 · Cited by 12 · see citing works
Vocal signals of ontogeny and fledging in nestling black-cockatoos: Implications for monitoring
Teixeira, D., Hill, R., Barth, M., Maron, M., & van Rensburg, B. J. (2021) · Bioacoustics · doi:10.1080/09524622.2021.1941257 · Cited by 10 · see citing works
A baseline survey of birds in native vegetation on cotton farms in inland eastern Australia
Smith, R., Reid, J. R. W., Scott-Morales, L. M., Green, S., & Reid, N. C. H. (2019) · Wildlife Research · doi:10.1071/wr18038 · Cited by 4 · see citing works
Flagged first, then by citations
Research works printed but not linked: 1
- Stowell D. Giannoulis E. Benetos M. Lagrange and M. D. Plumble (Piscataway, NJ: IEEE).
Cited by
Sorted by citations
Passive acoustic monitoring in terrestrial vertebrates: a review
Hoefer, S., McKnight, D. T., Allen‐Ankins, S., Nordberg, E. J., & Schwarzkopf, L. (2023) · Bioacoustics · doi:10.1080/09524622.2023.2209052 · Cited by 84 · see citing works
A Survey on Artificial Intelligence-Based Acoustic Source Identification
Zaheer, R., Ahmad, I., Habibi, D., Islam, K. Y., & Phung, Q. V. (2023) · IEEE Access · doi:10.1109/access.2023.3283982 · Cited by 42 · see citing works
The use of BirdNET embeddings as a fast solution to find novel sound classes in audio recordings
Allen‐Ankins, S., Hoefer, S. H., Bartholomew, J. I., Brodie, S., & Schwarzkopf, L. (2025) · Frontiers in Ecology and Evolution · doi:10.3389/fevo.2024.1409407 · Cited by 22 · see citing works
Rapid detection of microfibres in environmental samples using open-source visual recognition models
Galata, S., Walkington, I., Lane, T., Kiriakoulakis, K., & Dick, J. (2024) · Journal of Hazardous Materials · doi:10.1016/j.jhazmat.2024.135956 · Cited by 21 · see citing works
Kumar, S. V. S., & Kondaveeti, H. K. (2024) · Ecological Informatics · doi:10.1016/j.ecoinf.2024.102510 · Cited by 20 · see citing works
HawkEars: A regional, high-performance avian acoustic classifier
Huus, J., Kelly, K. G., Bayne, E. M., & Knight, E. C. (2025) · Ecological Informatics · doi:10.1016/j.ecoinf.2025.103122 · Cited by 16 · see citing works
Bird Species Detection Using CNN and EfficientNet-B0
Neeli, S., Guruguri, C. S. R., Kammara, A. R. A., Annepu, V., Bagadi, K., & Chirra, V. R. R. (2023) · International Conference on Next Generation Electronics (NELEX) · doi:10.1109/nelex59773.2023.10420966 · Cited by 9 · see citing works
Doohan, B., Hoefer, S. H., Allen‐Ankins, S., Nilsen, V., & Schwarzkopf, L. (2025) · Ecological Indicators · doi:10.1016/j.ecolind.2025.114533 · Cited by 8 · see citing works
Wang, J., & Tabeta, S. (2023) · Ecological Informatics · doi:10.1016/j.ecoinf.2023.102321 · Cited by 6 · see citing works
Sound-based bird classification using multiple features and machine learning paradigms
Revathi, A. A., & Sasikaladevi, N. (2025) · Multimedia Tools and Applications · doi:10.1007/s11042-024-20565-5 · Cited by 6 · see citing works
A deep learning approach to detect and identify live freshwater macroinvertebrates
Jaballah, S., Garcia, G. F., Martignac, F., Parisey, N., Jumel, S., Roussel, J., & Dézerald, O. (2023) · Aquatic Ecology · doi:10.1007/s10452-023-10053-7 · Cited by 5 · see citing works
Alexander, C., Clemens, R. S., Roe, P., & Fuller, S. J. (2025) · Ecological Informatics · doi:10.1016/j.ecoinf.2025.103222 · Cited by 4 · see citing works
Self-citationPatterns of Vocal Activity of the Chinese Bamboo Partridge Using BirdNET Analyzer
Mei, J., Li, L., Zhang, W., Shi, J., Zhao, S., Yong, F., Ge, X., Tong, W., Zhou, X. S., & Cui, P. L. (2026) · Animals · doi:10.3390/ani16020303 · Cited by 4 · see citing works
Robust sound-based bird classification using multiple features and random forest classifier
Revathi, A., & Sasikaladevi, N. (2025) · International Journal of Speech Technology · doi:10.1007/s10772-025-10168-2 · Cited by 5 · see citing works
Towards automated detection of the endangered southern black-throated finch (Poephila cincta cincta)
Allen‐Ankins, S., Laguna, J. M., & Schwarzkopf, L. (2024) · Wildlife Research · doi:10.1071/wr23151 · Cited by 3 · see citing works
An Optimised Grid Search Based Framework for Robust Large-Scale Natural Soundscape Classification
Napier, T., Ahn, E., Allen‐Ankins, S., & Lee, I. (2023) · Lecture notes in computer science · doi:10.1007/978-981-99-8388-9_38 · Cited by 3 · see citing works
Data-Centric Methods for Environmental Sound Classification With Limited Labels
Syed, A. R., Çoban, E. B., Pir, D., & Mandel, M. (2024) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2024.3414332 · Cited by 3 · see citing works
Xie, J., Xie, S., Liu, Y., Jing, X., Zhu, M., Xie, L., Zhang, J., Qian, K., & Schuller, B. W. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11414-4 · Cited by 3 · see citing works
BirdNET: Automated Detection for Monitoring Critically Endangered Lemurs from the Maromizaha Forest
Ferrario, V., Versaci, G., Dall'Ava, G., Gregorio, C. D., Carugati, F., Cristiano, W., Torti, V., Ratsimbazafy, J., Giacoma, C., Gamba, M., et al. (2026) · Integrative Zoology · doi:10.1111/1749-4877.70094 · Cited by 2 · see citing works
Sharma, B., Kumar, A., Singh, R., & Lehana, P. K. (2026) · Multimedia Tools and Applications · doi:10.1007/s11042-026-21646-3 · Cited by 2 · see citing works
Azhari, M. F., Syah, G. M., Nurdiati, S., Khatizah, E., & Najib, M. K. (2025) · Turkish Journal of Mathematics and Computer Science · doi:10.47000/tjmcs.1689564 · Cited by 1 · see citing works
Roars, Rumbles, and Resonance: A Systematic Review and Meta‐Analysis of Crocodylian Acoustic Signals
Flores, S. A., Dwyer, R. G., Parsons, S., & Potvin, D. A. (2026) · Ecology and Evolution · doi:10.1002/ece3.72494
Self-citationAcoustic individual identification in a species of field cricket using deep learning
Kabuga, E., Nandi, D., Burrell, S. A., Dlamini, G. S., Balakrishnan, R., Bah, B., & Durbach, I. N. (2026) · The Journal of the Acoustical Society of America · doi:10.1121/10.0044100
Aquatic biodiversity along the Danube River
Mikuška, T., Cogălniceanu, D., Dokulil, M. T., Freyhof, J., Makovinská, J., Pall, K., Paunović, M., Strefke, K., & Teubner, K. (2025) · Elsevier eBooks · doi:10.1016/b978-0-443-18686-8.00019-6
Huber, K., Waddle, J. H., Glorioso, B. M., & Donovan, T. M. (2026) · Ecological Informatics · doi:10.1016/j.ecoinf.2026.103764
Passive Acoustic Identification of Social Groups in the Hainan Gibbon
Kabuga, E., Dufourq, E., Luphade, L. N., Turvey, S., Ma, H., Cheyne, S. M., Lui, H., Bryant, J. V., Chen, Q., Li, W., et al. (2026) · Remote Sensing in Ecology and Conservation · doi:10.1002/rse2.70088
Kontogiannis, S. I., Arfanis, G., Biniaris, J., Bounas, A., & Kokkonis, G. (2026) · Electronics · doi:10.3390/electronics15204724
Two decades of machine learning in bioacoustics: a systematic review
Mora-González, E., Canepa, A., Aparicio, J. A. B., & González, Á. A. (2026) · Applied Intelligence · doi:10.1007/s10489-026-07454-0
Listening to the City: Evaluating BirdNET Reliability and Bias in Urban Bird Community Assessments
Ferrario, V., Alba, R., Marcolin, F., Brambilla, G., Chamberlain, D., & Caprio, E. (2026) · Remote Sensing in Ecology and Conservation · doi:10.1002/rse2.70116
- Authors
- Eichinski, P., Alexander, C., Roe, P., Parsons, S., & Fuller, S.
- Affiliations
- Queensland University of Technology (AU)
- Corresponding author
- Eichinski, Philip
- Venue
- Frontiers in Ecology and Evolution
- Publisher
- Frontiers Media SA
- Published
- 2022-03-14
- Volume
- 10
- ISSN
- 2296-701X (ISSN-L)
- OpenAlex
- W4220799428
- Cited by
- 29
- References
- 25
- Citations by year
- 2026: 10 · 2025: 9 · 2024: 4 · 2023: 6
- FWCI
- 2.19
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Frontiers in Ecology and Evolution (publishedVersion, cc-by-sa)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Cotton Research and Development Corporation
- Topics
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Life Sciences › Biochemistry, Genetics and Molecular Biology › Developmental Biology › Animal Vocal Communication and Behavior
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Keywords
- bird identification, ecoacoustics, convolutional neural networks, limited training data, data mismatch, iterative training, semi-supervised learning, cotton-growing zone, audio classification
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
