Where are we in semantic concept extraction for Spoken Language Understanding?

TL;DR

This paper reviews recent advances in semantic concept extraction for Spoken Language Understanding, highlighting the shift to end-to-end neural models and demonstrating significant performance improvements on the French MEDIA benchmark.

Contribution

It provides an overview of recent progress in SLU, especially in end-to-end neural approaches, and reports new results that outperform current state-of-the-art systems.

Findings

End-to-end neural models outperform cascade approaches in SLU.

Self-supervised training enhances speech and semantic extraction.

Achieved a Concept Error Rate of 11.2%, surpassing previous best.

Abstract

Spoken language understanding (SLU) topic has seen a lot of progress these last three years, with the emergence of end-to-end neural approaches. Spoken language understanding refers to natural language processing tasks related to semantic extraction from speech signal, like named entity recognition from speech or slot filling task in a context of human-machine dialogue. Classically, SLU tasks were processed through a cascade approach that consists in applying, firstly, an automatic speech recognition process, followed by a natural language processing module applied to the automatic transcriptions. These three last years, end-to-end neural approaches, based on deep neural networks, have been proposed in order to directly extract the semantics from speech signal, by using a single neural model. More recent works on self-supervised training with unlabeled data open new perspectives in term of performance for automatic speech recognition and natural language processing. In this paper, we present a brief overview of the recent advances on the French MEDIA benchmark dataset for SLU, with or without the use of additional data. We also present our last results that significantly outperform the current state-of-the-art with a Concept Error Rate (CER) of 11.2%, instead of 13.6% for the last state-of-the-art system presented this year.