Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.10074 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866918172031975424 |
|---|---|
| author | Sgouropoulos, Christos Nikou, Christos Vlachos, Stefanos Theiou, Vasileios Foukanelis, Christos Giannakopoulos, Theodoros |
| author_facet | Sgouropoulos, Christos Nikou, Christos Vlachos, Stefanos Theiou, Vasileios Foukanelis, Christos Giannakopoulos, Theodoros |
| contents | Few-shot learning has emerged as a powerful paradigm for training models with limited labeled data, addressing challenges in scenarios where large-scale annotation is impractical. While extensive research has been conducted in the image domain, few-shot learning in audio classification remains relatively underexplored. In this work, we investigate the effect of integrating supervised contrastive loss into prototypical few shot training for audio classification. In detail, we demonstrate that angular loss further improves the performance compared to the standard contrastive loss. Our method leverages SpecAugment followed by a self-attention mechanism to encapsulate diverse information of augmented input versions into one unified embedding. We evaluate our approach on MetaAudio, a benchmark including five datasets with predefined splits, standardized preprocessing, and a comprehensive set of few-shot learning models for comparison. The proposed approach achieves state-of-the-art performance in a 5-way, 5-shot setting. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_10074 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Prototypical Contrastive Learning For Improved Few-Shot Audio Classification Sgouropoulos, Christos Nikou, Christos Vlachos, Stefanos Theiou, Vasileios Foukanelis, Christos Giannakopoulos, Theodoros Sound Machine Learning Few-shot learning has emerged as a powerful paradigm for training models with limited labeled data, addressing challenges in scenarios where large-scale annotation is impractical. While extensive research has been conducted in the image domain, few-shot learning in audio classification remains relatively underexplored. In this work, we investigate the effect of integrating supervised contrastive loss into prototypical few shot training for audio classification. In detail, we demonstrate that angular loss further improves the performance compared to the standard contrastive loss. Our method leverages SpecAugment followed by a self-attention mechanism to encapsulate diverse information of augmented input versions into one unified embedding. We evaluate our approach on MetaAudio, a benchmark including five datasets with predefined splits, standardized preprocessing, and a comprehensive set of few-shot learning models for comparison. The proposed approach achieves state-of-the-art performance in a 5-way, 5-shot setting. |
| title | Prototypical Contrastive Learning For Improved Few-Shot Audio Classification |
| topic | Sound Machine Learning |
| url | https://arxiv.org/abs/2509.10074 |