Assessing the Potential of Deep Learning in the Acoustic Identification of Threatened Birds in the Pantanal

• Gabriel A. Correia UFMT
• Thiago M. Ventura UFMT

DOI: https://doi.org/10.5753/eniac.2025.12424

Abstract

This study investigates the feasibility of using machine learning techniques applied to bioacoustics for the automated identification of threatened bird species in the Pantanal biome. Bioacoustic data were collected from the Xeno-canto platform and used in conjunction with convolutional neural networks and Mel spectrograms. The results demonstrated satisfactory performance, with recall values varying among the selected species: Anodorhynchus hyacinthinus (0.94), Calidris canutus (0.93), Calidris pusilla (0.80), Crax fasciolata (0.70), and Harpia harpyja (0.89). It was concluded that the application of machine learning for detecting threatened bird species is technically feasible, depending on data quality and methodological adjustments.

References

de Oliveira, A. G., Ventura, T. M., Ganchev, T. D., Silva, L. N., Marques, M. I., and Schuchmann, K.-L. (2020). Speeding up training of automated bird recognizers by data reduction of audio features. PeerJ, 8:e8407.

Instituto Chico Mendes de Conservação da Biodiversidade (2018). Livro Vermelho da Fauna Brasileira Ameaçada de Extinção: Volume III – Aves, volume 3 of Livro Vermelho da Fauna Brasileira Ameaçada de Extinção. ICMBio/MMA, Brasília, DF, 1 edition. 7 volumes. Conteúdo: v.1. Geral; v.2. Mamíferos; v.3. Aves; v.4. Répteis; v.5. Anfíbios; v.6. Peixes; v.7. Invertebrados (Pencard). Ilustrado.

IUCN (2025). The iucn red list of threatened species. version 2025-1. [link]. Acesso: 10 de abril de 2025.

Junk, W. J., Da Cunha, C. N., Wantzen, K. M., Petermann, P., Strüssmann, C., Marques, M. I., and Adis, J. (2006). Biodiversity and its conservation in the pantanal of mato grosso, brazil. Aquatic Sciences, 68:278–309.

Kumar, A. S., Schlosser, T., Kahl, S., and Kowerko, D. (2024). Improving learning-based birdsong classification by utilizing combined audio augmentation strategies. Ecological Informatics, 82:102699.

Nunes, A. P., Posso, S. R., Frota, A. V. B. d., Vitorino, B. D., Laps, R. R., Donatelli, R. J., Straube, F. C., Pivatto, M. A. C., Oliveira, D. M. M. d., Carlos, B., and et al. (2021). Birds of the pantanal floodplains, brazil: historical data, diversity, and conservation. Papéis Avulsos de Zoologia, 61:e20216182.

Queensland University of Technology’s Ecoacoustics Research Group (2024). Eco-sounds. Acesso: 10 de abril de 2025.

Sharma, S., Sato, K., and Gautam, B. P. (2022). Bioacoustics monitoring of wildlife using artificial intelligence: A methodological literature review. In 2022 International Conference on Networking and Network Applications (NaNA). IEEE.

Stowell, D. (2022). Computational bioacoustics with deep learning: a review and roadmap. PeerJ, 10:e13152.

Ventura, T. M., Ganchev, T. D., Pérez-Granados, C., de Oliveira, A. G., de S. G. Pedroso, G., Marques, M. I., and and, K.-L. S. (2024). The importance of acoustic background modelling in cnn-based detection of the neotropical white-lored spinetail (aves, passeriformes, furnaridae). Bioacoustics, 33(2):103–121.

Xeno-canto Foundation and Naturalis Biodiversity Center (2024). xeno-canto: Sharing wildlife sounds from around the world. Acesso: 10 de abril de 2025.

Xie, J., Hu, K., Zhu, M., Yu, J., and Zhu, Q. (2019). Investigation of different cnn-based models for improved bird sound classification. IEEE Access, 7:175353–175361.