Training neural audio classifiers with few data

Pons Puig, Jordi; Serrà Julià, Joan; Serra, Xavier

doi:http://dx.doi.org/10.1109/ICASSP.2019.8682591

Training neural audio classifiers with few data

Citation

Pons J, Serrà J, Serra X. Training neural audio classifiers with few data. In: 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP); 2019 May 12-17; Brighton, United Kingdom. New Jersey: Institute of Electrical and Electronics Engineers; 2019. p. 16-20. DOI: 10.1109/ICASSP.2019.8682591

Abstract

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.

Description

Comunicació presentada a: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing celebrat de 12 al 17 de maig de 2019 a Brighton, Regne Unit.