Smart Passive Acoustic Monitoring: Embedding a Classifier on AudioMoth Microcontroller

TL;DR

This paper presents a resource-efficient embedded classifier for AudioMoth microcontrollers to enable real-time, in-situ bioacoustic monitoring of endangered seabirds, improving data collection and analysis efficiency.

Contribution

It introduces an optimized 1D-CNN model embedded on AudioMoth, enabling real-time classification with minimal resources and providing an open-source tutorial for model optimization.

Findings

Achieved 91% classification accuracy on seabird calls.

Reduced model size to ~10kB RAM footprint and 20ms inference time.

Enhanced AudioMoth firmware with real-time classification and selective recording functions.

Abstract

Passive Acoustic Monitoring (PAM) is an efficient and non-invasive method for surveying ecosystems at a reduced cost. Typically, autonomous recorders allow the acquisition of vast bioacoustic datasets which are then analyzed. However, power consumption and data storage are both scarce and limit the duration of acquisition campaigns. To address this issue, we propose a smart PAM system which allows the in-situ analysis of the soundscape by embedding a classifier directly onto an AudioMoth microcontroller. Specifically, we propose an optimized yet simple 1D Convolutional Neural Network (1D-CNN) to classify the raw audio. The model focuses on the specific call of Scopoli Shearwater seabirds (endangered species) and is trained on a real-world dataset with a classification accuracy of 91\% (balanced accuracy of 89\%). We also propose a process to optimize the model to fit the severe resource constraints of the AudioMoth, achieving a \~10kB RAM memory footprint and 20ms inference time. Finally, we present an open-source tutorial of our model optimization and export strategy which can be used for embedding models beyond the scope of our study. Our modified version of the AudioMoth firmware adds two functions: (F1) which selectively records data when the target species has been detected and (F2) which logs the continuous classification results in real time. This work intends to facilitate the conception of intelligent sensors, enhancing the efficiency and scalability of bioacoustic monitoring campaigns.