Balanced Deep CCA for Bird Vocalization Detection

TL;DR

This paper introduces balanced DCCA, a self-supervised learning method that enhances multi-modal bird vocalization detection by addressing data imbalance and event sparseness, leading to improved embedding quality.

Contribution

It proposes a novel balanced DCCA technique that overcomes event sparseness and label imbalance in multi-modal data for better detection performance.

Findings

Balanced DCCA outperforms classical DCCA in detection tasks.

The method effectively handles label imbalance in low-resource scenarios.

Improved embeddings lead to higher detection accuracy.

Abstract

Event detection improves when events are captured by two different modalities rather than just one. But to train detection systems on multiple modalities is challenging, in particular when there is abundance of unlabelled data but limited amounts of labeled data. We develop a novel self-supervised learning technique for multi-modal data that learns (hidden) correlations between simultaneously recorded microphone (sound) signals and accelerometer (body vibration) signals. The key objective of this work is to learn useful embeddings associated with high performance in downstream event detection tasks when labeled data is scarce and the audio events of interest (songbird vocalizations) are sparse. We base our approach on deep canonical correlation analysis (DCCA) that suffers from event sparseness. We overcome the sparseness of positive labels by first learning a data sampling model from the labelled data and by applying DCCA on the output it produces. This method that we term balanced DCCA (b-DCCA) improves the performance of the unsupervised embeddings on the downstream supervised audio detection task compared to classsical DCCA. Because data labels are frequently imbalanced, our method might be of broad utility in low-resource scenarios.