# Training neural audio classifiers with few data

arXiv: https://arxiv.org/abs/1810.10274 · HTML: https://mlsanity.com/paper/1810.10274/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 24 Oct 2018.

## Built on by (most cited first, 9 of 9)

- [Adaptive Few-Shot Learning Algorithm for Rare Sound Event Detection](https://mlsanity.com/paper/2205.11738/) (2022)
- [HalluAudio: Hallucinating Frequency as Concepts for Few-Shot Audio Classification](https://mlsanity.com/paper/2302.14204/) (2023)
- [Meta-Learning in Audio and Speech Processing: An End to End Comprehensive Review](https://mlsanity.com/paper/2408.10330/) (2024)
- [Prototypical Contrastive Learning For Improved Few-Shot Audio Classification](https://mlsanity.com/paper/2509.10074/) (2025)
- [One-shot learning for acoustic identification of bird species in non-stationary environments](https://mlsanity.com/paper/2105.00202/) (2021)
- [Structural and Statistical Audio Texture Knowledge Distillation for Acoustic Classification](https://mlsanity.com/paper/2501.01921/) (2025)
- [MetaCLBench: Meta Continual Learning Benchmark on Resource-Constrained Edge Devices](https://mlsanity.com/paper/2504.00174/) (2025)
- [A Comprehensive Survey of Challenges and Opportunities of Few-Shot Learning Across Multiple Domains](https://mlsanity.com/paper/2504.04017/) (2025)
- [Learning Music Audio Representations With Limited Data](https://mlsanity.com/paper/2505.06042/) (2025)

## Abstract (arXiv)

We investigate supervised learning strategies that improve the training of neural network audio classifiers on small annotated collections. In particular, we study whether (i) a naive regularization of the solution space, (ii) prototypical networks, (iii) transfer learning, or (iv) their combination, can foster deep learning models to better leverage a small amount of training examples. To this end, we evaluate (i-iv) for the tasks of acoustic event recognition and acoustic scene classification, considering from 1 to 100 labeled examples per class. Results indicate that transfer learning is a powerful strategy in such scenarios, but prototypical networks show promising results when one does not count with external or validation data.