A Sequential Self Teaching Approach for Improving Generalization in Sound Event Recognition

Abstract

An important problem in machine auditory perception is to recognize and detect sound events. In this paper, we propose a sequential self-teaching approach to learning sounds. Our main proposition is that it is harder to learn sounds in adverse situations such as from weakly labeled and/or noisy labeled data, and in these situations a single stage of learning is not sufficient. Our proposal is a sequential stage-wise learning process that improves generalization capabilities of a given modeling system. We justify this method via technical results and on Audioset, the largest sound events dataset, our sequential learning approach can lead to up to 9% improvement in performance. A comprehensive evaluation also shows that the method leads to improved transferability of knowledge from previously trained models, thereby leading to improved generalization capabilities on transfer learning tasks.

Accepted International Conference on Machine Learning $\text{(ICML) 2020}$. 14 pages

Cited by in corpus (7)

• An Ensemble of Convolutional Neural Networks for Audio Classification
• Pruning vs XNOR-Net: A Comprehensive Study of Deep Learning for Audio Classification on Edge-devices
• NORESQA: A Framework for Speech Quality Assessment using Non-Matching References
• Study of positional encoding approaches for Audio Spectrogram Transformers
• GISE-51: A scalable isolated sound events dataset
• Conditional Teaching Size
• Incorporating Real-world Noisy Speech in Neural-network-based Speech Enhancement Systems