BioME: A Resource-Efficient Bioacoustic Foundational Model for IoT Applications

2026-02-10

AI Summary

This work addresses the high computational cost and poor generalization of existing self-supervised audio encoders in bioacoustic tasks, which hinder their deployment on resource-constrained edge devices. The authors propose a lightweight encoder that compresses parameters from a high-capacity teacher model via inter-layer knowledge distillation and enhances ecological generalization through multi-domain self-supervised pretraining on speech, environmental sounds, and animal vocalizations. A novel integration of FiLM-based conditional modulation injects digital signal processing priors to improve feature disentanglement in the low-capacity student model. With only 25% of the original parameter count, the proposed model matches or exceeds the performance of the full-scale counterpart across multiple bioacoustic benchmarks, substantially reducing computational demands while remaining suitable for IoT edge deployment.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Natural Language Processing: Learning & Optimization for NLP

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Passive acoustic monitoring has become a key strategy in biodiversity assessment, conservation, and behavioral ecology, especially as Internet-of-Things (IoT) devices enable continuous in situ audio collection at scale. While recent self-supervised learning (SSL)-based audio encoders, such as BEATs and AVES, have shown strong performance in bioacoustic tasks, their computational cost and limited robustness to unseen environments hinder deployment on resource-constrained platforms. In this work, we introduce BioME, a resource-efficient audio encoder designed for bioacoustic applications. BioME is trained via layer-to-layer distillation from a high-capacity teacher model, enabling strong representational transfer while reducing the parameter count by 75%. To further improve ecological generalization, the model is pretrained on multi-domain data spanning speech, environmental sounds, and animal vocalizations. A key contribution is the integration of modulation-aware acoustic features via FiLM conditioning, injecting a DSP-inspired inductive bias that enhances feature disentanglement in low-capacity regimes. Across multiple bioacoustic tasks, BioME matches or surpasses the performance of larger models, including its teacher, while being suitable for resource-constrained IoT deployments. For reproducibility, code and pretrained checkpoints are publicly available.

Problem

• bioacoustics
• resource-constrained
• IoT
• self-supervised learning
• model deployment

Innovation

• layer-to-layer distillation
• modulation-aware features
• FiLM conditioning
• resource-efficient audio encoder
• multi-domain pretraining