Lightweight Convolutional Neural Network for Robust  and Label-Efficient Animal Sound Classification

Ivan Saputra; Rikman Aherliwan Rudawan

doi:10.29207/resti.v10i4.7366

Authors

Ivan Saputra Telkom University https://orcid.org/0009-0004-7668-0898
Rikman Aherliwan Rudawan Telkom University https://orcid.org/0000-0003-3169-2663

DOI:

https://doi.org/10.29207/resti.v10i4.7366

Keywords:

animal sound classification, bioacoustic monitoring, label efficiency, lightweight CNN, Log-Mel spectrogram

Abstract

Automated bioacoustic monitoring is increasingly important for biodiversity observation, yet practical deployment is often limited by computational constraints and the scarcity of annotated audio data. This study introduces MelCNN, a custom compact end-to-end architecture designed to mitigate over-parameterization and negative transfer in animal sound classification under small-data conditions. A balanced dataset of 600 audio clips from three animal classes (Cat, Dog, and Cow) was standardized to 4.8 seconds, converted into Log-Mel spectrograms, and evaluated using 5-fold stratified cross-validation. The proposed Lightweight CNN with Global Average Pooling was compared against YAMNet-based transfer learning baselines and three additional modern lightweight CNN baselines, namely MobileNetV3-Small, ShuffleNetV2-1.0x, and EfficientNet-Lite0-style. The proposed model achieved 97.00% ± 2.01% Accuracy and 97.01% ± 2.00% Macro-F1 while retaining a compact parameter size of approximately 0.39 million, outperforming both the strongest transfer learning baselines and the added modern lightweight baselines in the present setting. The model also maintained performance above 90% at 12 kHz and preserved approximately 90.5% Macro-F1 when trained with only 30% of the available labeled data. These findings indicate that a domain-specific lightweight architecture can provide a favorable accuracy-efficiency trade-off for controlled small-data animal sound classification in resource-constrained monitoring scenarios.

Downloads

Download data is not yet available.

Author Biography

Ivan Saputra, Telkom University

Telkom University Student | Multimedia Technology Enthusiast | Eager to Take on New Challenges

References

C. A. Quinn et al., “Soundscape classification with convolutional neural networks reveals temporal and geographic patterns in ecoacoustic data,” Ecol. Indic., vol. 138, p. 108831, May 2022, doi: https://doi.org/10.1016/j.ecolind.2022.108831

F. Sessa et al., “Advancing Diagnostic Tools in Forensic Science: The Role of Artificial Intelligence in Gunshot Wound Investigation—A Systematic Review,” Forensic Sciences, vol. 5, no. 3, p. 30, Sep. 2025, doi: https://doi.org/10.3390/forensicsci5030030

J. Zurn, W. Burgard, and A. Valada, “Self-Supervised Visual Terrain Classification from Unsupervised Acoustic Feature Learning,” IEEE Transactions on Robotics, vol. 37, no. 2, pp. 466–481, Apr. 2021, doi: https://doi.org/10.1109/TRO.2020.3031214

B. Swaminathan, M. Jagadeesh, and S. Vairavasundaram, “Multi-label classification for acoustic bird species detection using transfer learning approach,” Ecol. Inform., vol. 80, p. 102471, May 2024, doi: https://doi.org/10.1016/j.ecoinf.2024.102471

H. Xu, Y. Tian, H. Ren, and X. Liu, “A Lightweight Channel and Time Attention Enhanced 1D CNN Model for Environmental Sound Classification,” Expert Syst. Appl., vol. 249, p. 123768, Sep. 2024, doi: https://doi.org/10.1016/j.eswa.2024.123768

M. K. H. Kanchon, M. Sadman, K. F. Nabila, R. Tarannum, and R. Khan, “Enhancing personalized learning: AI-driven identification of learning styles and content modification strategies,” International Journal of Cognitive Computing in Engineering, vol. 5, pp. 269–278, Jan. 2024, doi: https://doi.org/10.1016/j.ijcce.2024.06.002

B. Tiple and M. Patwardhan, “Multi-label emotion recognition from Indian classical music using gradient descent SNN model,” Multimed. Tools Appl., vol. 81, no. 6, pp. 8853–8870, Mar. 2022, doi: https://doi.org/10.1007/s11042-022-11975-4

Y. Chen et al., “Effective Audio Classification Network Based on Paired Inverse Pyramid Structure and Dense MLP Block,” in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer Science and Business Media Deutschland GmbH, 2023, pp. 70–84, doi: https://doi.org/10.1007/978-981-99-4742-3_6

Z. Fang, B. Yin, Z. Du, and X. Huang, “Fast environmental sound classification based on resource adaptive convolutional neural network,” Sci. Rep., vol. 12, no. 1, p. 10382, Dec. 2022, doi: https://doi.org/10.1038/s41598-022-10382-x

J. Park, T. Yoo, S. Lee, and T. Kim, “Urban Noise Analysis and Emergency Detection System using Lightweight End-to-End Convolutional Neural Network,” International Journal of Computers, Communications and Control, vol. 18, no. 5, p. 5814, Oct. 2023, doi: https://doi.org/10.15837/ijccc.2023.5.5814

A. H. Z. Ilahi, A. Irwansyah, and H. Oktavianto, “Comparative Study of CNN Architectures for Real-Time Audio-Based Car Accident Detection on Edge Devices,” JOIV : International Journal on Informatics Visualization, vol. 9, no. 3, p. 1310, May 2025, doi: https://doi.org/10.62527/joiv.9.3.2985

A. Khodadadi, S. Molaei, M. Teimouri, and H. Zare, “Classification of audio codecs with variable bit-rates using deep-learning methods,” Digital Signal Processing: A Review Journal, vol. 110, p. 102952, Mar. 2021, doi: https://doi.org/10.1016/j.dsp.2020.102952

S. Anand, N. K. Devulapally, S. Das Bhattacharjee, and J. Yuan, “Multi-label Emotion Analysis in Conversation via Multimodal Knowledge Distillation,” in MM 2023 - Proceedings of the 31st ACM International Conference on Multimedia, Association for Computing Machinery, Inc, Oct. 2023, pp. 6090–6100, doi: https://doi.org/10.1145/3581783.3612517

A. M. Tripathi and O. J. Pandey, “Divide and Distill: New Outlooks on Knowledge Distillation for Environmental Sound Classification,” IEEE/ACM Trans. Audio Speech Lang. Process., vol. 31, pp. 1100–1113, 2023, doi: https://doi.org/10.1109/TASLP.2023.3244507

C. Fan, J. Hu, and J. Huang, “Few-Shot Multi-Agent Perception,” in MM 2021 - Proceedings of the 29th ACM International Conference on Multimedia, Association for Computing Machinery, Inc, Oct. 2021, pp. 1712–1720, doi: https://doi.org/10.1145/3474085.3475315

T. Likhomanenko, Q. Xu, J. Kahn, G. Synnaeve, and R. Collobert, “slimIPL: Language-model-free iterative pseudo-labeling,” in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, International Speech Communication Association, 2021, pp. 2118–2122, doi: https://doi.org/10.21437/Interspeech.2021-740

J. Liang et al., “Adapting Language-Audio Models as Few-Shot Audio Learners,” in Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH, International Speech Communication Association, 2023, pp. 276–280, doi: https://doi.org/10.21437/Interspeech.2023-1082

K. M. Hasib, A. Tanzim, J. Shin, K. O. Faruk, J. Al Mahmud, and M. F. Mridha, “BMNet-5: A Novel Approach of Neural Network to Classify the Genre of Bengali Music Based on Audio Features,” IEEE Access, vol. 10, pp. 108545–108563, 2022, doi: https://doi.org/10.1109/ACCESS.2022.3213818

Y. Wang, C. Liu, H. Wu, Q. Sui, C. Yang, and W. Gui, “Revolutionizing Flotation Process Working Condition Identification Based on Froth Audio,” IEEE Trans. Instrum. Meas., vol. 72, p. 3315388, 2023, doi: https://doi.org/10.1109/TIM.2023.3315388

J. Segura-Garcia, S. Sturley, M. Arevalillo-Herraez, J. M. Alcaraz-Calero, S. Felici-Castell, and E. A. Navarro-Camba, “5G AI-IoT System for Bird Species Monitoring and Song Classification,” Sensors, vol. 24, no. 11, p. 3687, Jun. 2024, doi: https://doi.org/10.3390/s24113687

K. Shaheed et al., “EfficientRMT-Net—An Efficient ResNet-50 and Vision Transformers Approach for Classifying Potato Plant Leaf Diseases,” Sensors, vol. 23, no. 23, p. 9516, Dec. 2023, doi: https://doi.org/10.3390/s23239516

H. Bakır, A. N. Çayır, and T. S. Navruz, “A comprehensive experimental study for analyzing the effects of data augmentation techniques on voice classification,” Multimed. Tools Appl., vol. 83, no. 6, pp. 17601–17628, Feb. 2024, doi: https://doi.org/10.1007/s11042-023-16200-4

N. Zacarias-Morales, J. A. Hernández-Nolasco, and P. Pancardo, “Full single-type deep learning models with multihead attention for speech enhancement,” Applied Intelligence, vol. 53, no. 17, pp. 20561–20576, Sep. 2023, doi: https://doi.org/10.1007/s10489-023-04571-y

W. Wang, Q. Li, J. Xie, N. Hu, Z. Wang, and N. Zhang, “Research on emotional semantic retrieval of attention mechanism oriented to audio-visual synesthesia,” Neurocomputing, vol. 519, pp. 194–204, Jan. 2023, doi: https://doi.org/10.1016/j.neucom.2022.11.036

Z. Zhao, E. B. A. Bakar, N. B. A. Razak, and M. N. Akhtar, “Corrosion image classification method based on EfficientNetV2,” Heliyon, vol. 10, no. 17, p. e36754, Sep. 2024, doi: https://doi.org/10.1016/j.heliyon.2024.e36754

V. Naresh, G. Yogeswararao, B. Nageswararao Naik, R. Malmathanraj, and P. Palanisamy, “Empirical Analysis of Squeeze and Excitation-Based Densely Connected CNN for Chili Leaf Disease Identification,” IEEE Transactions on Artificial Intelligence, vol. 5, no. 4, pp. 1681–1692, Apr. 2024, doi: https://doi.org/10.1109/TAI.2024.3364126

K. Filus and J. Domańska, “Global Entropy Pooling layer for Convolutional Neural Networks,” Neurocomputing, vol. 555, p. 126615, Oct. 2023, doi: https://doi.org/10.1016/j.neucom.2023.126615

T. B. Sridharan and D. P. S. Sharagaharajan Akilashri, “Multimodal learning analytics for students behavior prediction using multi-scale dilated deep temporal convolution network with improved chameleon Swarm algorithm,” Expert Syst. Appl., vol. 286, p. 128113, Aug. 2025, doi: https://doi.org/10.1016/j.eswa.2025.128113