Alignement automatique pour la compréhension littérale de l'oral par approche segmentale

Abstract : Most recent efficient statistical approaches for language understanding require a segmental annotation of the training data. In this paper we study an alternative that obtains a segmental alignment of conceptual units with words in an unsupervised way. The impact of the automatic alignment on the understanding system performance is evaluated on a spoken dialogue task.
Document type :
Conference papers
Complete list of metadatas

Cited literature [13 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01317556
Contributor : Bibliothèque Universitaire Déposants Hal-Avignon <>
Submitted on : Monday, February 18, 2019 - 10:51:36 AM
Last modification on : Wednesday, May 15, 2019 - 10:12:03 AM
Long-term archiving on : Sunday, May 19, 2019 - 5:27:48 PM

File

TALN11b.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01317556, version 1

Collections

Citation

Stéphane Huet, Fabrice Lefèvre. Alignement automatique pour la compréhension littérale de l'oral par approche segmentale. 18ème conférence sur le Traitement Automatique des Langues Naturelles (TALN), Jun 2011, Montpellier, France. ⟨hal-01317556⟩

Share

Metrics

Record views

59

Files downloads

14