Automatic acoustic identification of individual animals: Improving generalisation across species and recording conditions

TL;DR

This paper presents a general automatic method for identifying individual animals by their vocal sounds, capable of working across species and conditions, and introduces new analysis techniques to evaluate and improve robustness.

Contribution

The study introduces a versatile identification method applicable across multiple species and communication complexities, along with dataset manipulation techniques to assess and enhance classifier robustness.

Findings

Confirmed presence of confounds in previous studies

Proposed data manipulations to reduce confounds

Advocated for dataset sharing and standard confound assessment

Abstract

Many animals emit vocal sounds which, independently from the sounds' function, embed some individually-distinctive signature. Thus the automatic recognition of individuals by sound is a potentially powerful tool for zoology and ecology research and practical monitoring. Here we present a general automatic identification method, that can work across multiple animal species with various levels of complexity in their communication systems. We further introduce new analysis techniques based on dataset manipulations that can evaluate the robustness and generality of a classifier. By using these techniques we confirmed the presence of experimental confounds in situations resembling those from past studies. We introduce data manipulations that can reduce the impact of these confounds, compatible with any classifier. We suggest that assessment of confounds should become a standard part of future studies to ensure they do not report over-optimistic results. We provide annotated recordings used for analyses along with this study and we call for dataset sharing to be a common practice to enhance development of methods and comparisons of results.