Representation based meta-learning for few-shot spoken intent recognition

TL;DR

This paper introduces a meta-learning approach for few-shot spoken intent recognition that creates task-agnostic representations, enabling accurate classification with minimal training data on popular datasets.

Contribution

It proposes a novel representation-based meta-learning framework for few-shot spoken intent detection, demonstrating competitive performance with limited training samples.

Findings

Achieves 88.6% accuracy in 5-shot classification on Google Commands

Achieves 78.5% accuracy in 5-shot classification on Fluent Speech Commands

Performance comparable to traditional supervised models with more data

Abstract

Spoken intent detection has become a popular approach to interface with various smart devices with ease. However, such systems are limited to the preset list of intents-terms or commands, which restricts the quick customization of personal devices to new intents. This paper presents a few-shot spoken intent classification approach with task-agnostic representations via meta-learning paradigm. Specifically, we leverage the popular representation-based meta-learning learning to build a task-agnostic representation of utterances, that then use a linear classifier for prediction. We evaluate three such approaches on our novel experimental protocol developed on two popular spoken intent classification datasets: Google Commands and the Fluent Speech Commands dataset. For a 5-shot (1-shot) classification of novel classes, the proposed framework provides an average classification accuracy of 88.6% (76.3%) on the Google Commands dataset, and 78.5% (64.2%) on the Fluent Speech Commands dataset. The performance is comparable to traditionally supervised classification models with abundant training samples.