Synthetic Swarm Mosquito Dataset for Acoustic Classification: A Proof of Concept

2025-12-13

AI Summary

To address the high cost, limited scale, and narrow species coverage of real-world acoustic data collection in mosquito-borne disease surveillance, this paper proposes the first methodology for constructing a synthetic acoustic dataset of multi-species mosquito swarms under realistic mixed-species and noisy conditions. We establish a physics-informed, acoustics-driven paradigm for synthetic audio generation. Using log-mel spectrogram representations and lightweight deep models—including TinyCNN and MobileNetV2—we achieve end-to-end classification with high accuracy across six major vector mosquito species. The resulting models are optimized for embedded edge deployment, significantly reducing hardware power consumption and deployment costs. This work overcomes traditional data bottlenecks by providing a scalable, low-cost, real-time acoustic monitoring framework for mosquito vectors, accompanied by an open-source synthetic dataset and reproducible methodology.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Humans and AI: Crowd Sourcing and Human Computation
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Web Mining and Content Analysis: Web data generation and simulation
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search

Abstract

Mosquito-borne diseases pose a serious global health threat, causing over 700,000 deaths annually. This work introduces a proof-of-concept Synthetic Swarm Mosquito Dataset for Acoustic Classification, created to simulate realistic multi-species and noisy swarm conditions. Unlike conventional datasets that require labor-intensive recording of individual mosquitoes, the synthetic approach enables scalable data generation while reducing human resource demands. Using log-mel spectrograms, we evaluated lightweight deep learning architectures for the classification of mosquito species. Experiments show that these models can effectively identify six major mosquito vectors and are suitable for deployment on embedded low-power devices. The study demonstrates the potential of synthetic swarm audio datasets to accelerate acoustic mosquito research and enable scalable real-time surveillance solutions.

Problem

• Develop synthetic dataset for mosquito acoustic classification
• Enable scalable data generation to reduce manual recording
• Evaluate lightweight deep learning models for embedded deployment

Innovation

• Synthetic dataset simulates realistic multi-species noisy swarms
• Lightweight deep learning models classify six major mosquito vectors
• Scalable data generation enables deployment on low-power embedded devices