Weakly-Supervised Classification and Detection of Bird Sounds in the Wild. A BirdCLEF 2021 Solution

TL;DR

This paper presents a machine learning pipeline that detects and classifies bird sounds in complex natural soundscapes using weak labels, aiding ecological monitoring and biodiversity assessment.

Contribution

It introduces a novel weakly-supervised learning approach for bird sound detection and classification in wild environments, demonstrating robustness against background noise.

Findings

Achieved 10th place out of 816 teams in BirdCLEF 2021 Challenge

Effective detection of bird calls in noisy, real-world soundscapes

Robust classification of fine-grained bird vocalizations

Abstract

It is easier to hear birds than see them, however, they still play an essential role in nature and they are excellent indicators of deteriorating environmental quality and pollution. Recent advances in Machine Learning and Convolutional Neural Networks allow us to detect and classify bird sounds, by doing this, we can assist researchers in monitoring the status and trends of bird populations and biodiversity in ecosystems. We propose a sound detection and classification pipeline for analyzing complex soundscape recordings and identify birdcalls in the background. Our pipeline learns from weak labels, classifies fine-grained bird vocalizations in the wild, and is robust against background sounds (e.g., airplanes, rain, etc). Our solution achieved 10th place of 816 teams at the BirdCLEF 2021 Challenge hosted on Kaggle.