Augmented Contrastive Self-Supervised Learning for Audio Invariant Representations
This work addresses audio classification for applications like sound/event detection, but it appears incremental as it builds on existing self-supervised and contrastive learning techniques.
The paper tackles the challenge of generalization in audio classification due to labeled data scarcity by proposing an augmented contrastive self-supervised learning framework that learns invariant representations from unlabeled data, achieving significant performance improvements over state-of-the-art methods on Audioset and DESED datasets.
Improving generalization is a major challenge in audio classification due to labeled data scarcity. Self-supervised learning (SSL) methods tackle this by leveraging unlabeled data to learn useful features for downstream classification tasks. In this work, we propose an augmented contrastive SSL framework to learn invariant representations from unlabeled data. Our method applies various perturbations to the unlabeled input data and utilizes contrastive learning to learn representations robust to such perturbations. Experimental results on the Audioset and DESED datasets show that our framework significantly outperforms state-of-the-art SSL and supervised learning methods on sound/event classification tasks.