Machine Listening for Bioacoustics
My research interests lie in applying machine learning to automated sensing and perception of natural environments, particularly through sound. I aim to develop machine listening methods that help monitor the impacts humans have on ecosystems by enabling automated species detection and biodiversity estimation in bioacoustic data.
I am particularly interested in annotation-efficient machine listening methods for bioacoustics and ecoacoustics. I believe there is immense potential in using acoustic sensors and machine listening techniques to quantify the state of our natural environment, especially for monitoring animal populations.
The information in sound is often complementary to that of images, and the combination of the two can give a more complete understanding of the environment. While larger animals such as the fox may be easier to capture on camera, smaller but acoustically active animals like birds and frogs may be easier to capture with a microphone. I am excited about the possibilities of combining acoustic and visual sensor data to monitor ecosystems and the effects of human interventions.
Listen to This AI-Generated Podcast
A colleague of mine prompted NotebookLM with my previous webpage to generate a short podcast about my research vision. It offers an engaging listen and captures my aspirations for machine listening in bioacoustics quite well. Note that my contributions to the field are grossly exaggerated.
People I Work With
I am a PhD student at RISE working in the deep learning group, and affiliated with Lund University. My supervisors are Olof Mogren and Maria Sandsten. After a 2-month research visit at the audio research group at Tampere University, I have also been collaborating with Tuomas Virtanen.
Featured Publications
A selection of publications related to machine listening, bioacoustics, and environmental monitoring. You can find all my publications here.