An investigation of few-shot learning in spoken term classification

Chen，Yangbin1; Ko，Tom2; Shang，Lifeng3; Chen，Xiao3; Jiang，Xin3; Li，Qing4

10.21437/Interspeech.2020-2568

2020

2308-457X

1990-9772

Proceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH

2020-October

2582-2586

In this paper, we investigate the feasibility of applying few-shot learning algorithms to a speech task. We formulate a user-defined scenario of spoken term classification as a few-shot learning problem. In most few-shot learning studies, it is assumed that all the N classes are new in a N-way problem. We suggest that this assumption can be relaxed and define a N+Mway problem where N and M are the number of new classes and fixed classes respectively. We propose a modification to the Model-Agnostic Meta-Learning (MAML) algorithm to solve the problem. Experiments on the Google Speech Commands dataset show that our approach outperforms the conventional supervised learning approach and the original MAML.

Convolutional neural network Few-shot classification Meta learning Spoken term classification

通讯

英语

EI

20205209692001

Learning algorithms ; Speech communication

Machine Learning:723.4.2 ; Speech:751.5

2-s2.0-85098174751

Scopus

1.Department of Computer Science,City University of Hong Kong,Hong Kong
2.Department of Computer Science and Engineering,Southern University of Science and Technology,Shenzhen,China
3.Huawei Noah's Ark Lab,
4.Department of Computing,Hong Kong Polytechnic University,Hong Kong

Chen，Yangbin,Ko，Tom,Shang，Lifeng,et al. An investigation of few-shot learning in spoken term classification[C],2020:2582-2586.