RISE Research Radar

Computer Science Open House 2022-2025

2022

Machine learning and acoustic monitoring for wildlife conservation

John Martinsson

Summary

Research on automated analysis of years of audio data from natural habitats using machine learning. Focuses on few-shot learning for annotation-efficient methods and feature extraction for adapting time-frequency representations to acoustic models for wildlife sound detection.

Themes

machine-learningearth-observation

Keywords

machine listening, wildlife conservation, few-shot learning, acoustic monitoring, sound event detection, bioacoustics

Poster

Machine learning and acoustic monitoring for wildlife conservation poster

Click image to open full size