An Investigation of Few-Shot Learning in Spoken Term Classification
arXiv:1812.10233
Abstract
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+M-way 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.
arXiv commentAccepted by INTERSPEECH 2020
References in corpus (9)
- Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- Speech Commands: A Dataset for Limited-Vocabulary Speech Recognition
- A Simple Neural Attentive Meta-Learner
- Meta-SGD: Learning to Learn Quickly for Few-Shot Learning
- Hello Edge: Keyword Spotting on Microcontrollers
- End-to-End ASR-free Keyword Search from Speech
- Attention-based End-to-End Models for Small-Footprint Keyword Spotting
Cited by in corpus (6)
- Few-Shot Keyword Spotting With Prototypical Networks
- Representation based meta-learning for few-shot spoken intent recognition
- Towards Data-efficient Modeling for Wake Word Spotting
- Auto-KWS 2021 Challenge: Task, Datasets, and Baselines
- On Front-end Gain Invariant Modeling for Wake Word Spotting
- MetaMix: Improved Meta-Learning with Interpolation-based Consistency Regularization