# A Data Efficient End-To-End Spoken Language Understanding Architecture

arXiv: https://arxiv.org/abs/2002.05955 · HTML: https://mlsanity.com/paper/2002.05955/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 14 Feb 2020.

## Built on by (most cited first, 8 of 8)

- [LeBenchmark: A Reproducible Framework for Assessing Self-Supervised Representation Learning from Speech](https://mlsanity.com/paper/2104.11462/) (2021)
- [LeBenchmark 2.0: a Standardized, Replicable and Enhanced Framework for Self-supervised Representations of French Speech](https://mlsanity.com/paper/2309.05472/) (2023)
- [Rethinking End-to-End Evaluation of Decomposable Tasks: A Case Study on Spoken Language Understanding](https://mlsanity.com/paper/2106.15065/) (2021)
- [FANS: Fusing ASR and NLU for on-device SLU](https://mlsanity.com/paper/2111.00400/) (2021)
- [On the Use of Semantically-Aligned Speech Representations for Spoken Language Understanding](https://mlsanity.com/paper/2210.05291/) (2022)
- [ASR-Generated Text for Language Model Pre-training Applied to Speech Tasks](https://mlsanity.com/paper/2207.01893/) (2022)
- [Toward Low-Cost End-to-End Spoken Language Understanding](https://mlsanity.com/paper/2207.00352/) (2022)
- [Vers la compr'ehension automatique de la parole bout-en-bout `a moindre effort](https://mlsanity.com/paper/2207.00349/) (2022)

## Abstract (arXiv)

End-to-end architectures have been recently proposed for spoken language understanding (SLU) and semantic parsing. Based on a large amount of data, those models learn jointly acoustic and linguistic-sequential features. Such architectures give very good results in the context of domain, intent and slot detection, their application in a more complex semantic chunking and tagging task is less easy. For that, in many cases, models are combined with an external language model to enhance their performance. In this paper we introduce a data efficient system which is trained end-to-end, with no additional, pre-trained external module. One key feature of our approach is an incremental training procedure where acoustic, language and semantic models are trained sequentially one after the other. The proposed model has a reasonable size and achieves competitive results with respect to state-of-the-art while using a small training dataset. In particular, we reach 24.02% Concept Error Rate (CER) on MEDIA/test while training on MEDIA/train without any additional data.