Is it time to switch to Word Embedding and Recurrent Neural Networks for Spoken Language Understanding?

Vedran Vukotić 1 Christian Raymond 1 Guillaume Gravier 1
1 LinkMedia - Creating and exploiting explicit links between multimedia fragments
IRISA-D6 - MEDIA ET INTERACTIONS, Inria Rennes – Bretagne Atlantique
Abstract : Recently, word embedding representations have been investigated for slot filling in Spoken Language Understanding, along with the use of Neural Networks as classifiers. Neural Networks , especially Recurrent Neural Networks, that are specifically adapted to sequence labeling problems, have been applied successfully on the popular ATIS database. In this work, we make a comparison of this kind of models with the previously state-of-the-art Conditional Random Fields (CRF) classifier on a more challenging SLU database. We show that, despite efficient word representations used within these Neural Networks, their ability to process sequences is still significantly lower than for CRF, while also having a drawback of higher computational costs, and that the ability of CRF to model output label dependencies is crucial for SLU.
Document type :
Conference papers
Liste complète des métadonnées

Cited literature [19 references]  Display  Hide  Download

https://hal.inria.fr/hal-01196915
Contributor : Christian Raymond <>
Submitted on : Thursday, September 10, 2015 - 3:58:19 PM
Last modification on : Monday, December 17, 2018 - 9:06:01 AM
Document(s) archivé(s) le : Tuesday, December 29, 2015 - 12:06:16 AM

File

Interspeech2015.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01196915, version 1

Citation

Vedran Vukotić, Christian Raymond, Guillaume Gravier. Is it time to switch to Word Embedding and Recurrent Neural Networks for Spoken Language Understanding?. InterSpeech, Sep 2015, Dresde, Germany. ⟨hal-01196915⟩

Share

Metrics

Record views

1345

Files downloads

725