Summer Project: Identifying and analysing bat calls

Summary

Our goal is to use machine learning methods to identify patterns in large bat call data sets. Bat calls are recorded by a network of sensors that are set up in forests and near rivers. They record sounds at the frequencies of bat calls, producing typically noisy data. The first task of the project is to automatically identify bat calls in the data, and then find patterns in the identified calls and link it to specific behaviour. This research will help environmental researchers to better understand bat behaviour and distribution in NZ.