A corpus of spontaneous dialogues in L2 English by French and Japanese L1 speakers for automated assessment of fluency - Laboratoire de Linguistique et Didactique des Langues Etrangères et Maternelles
Poster De Conférence Année : 2024

A corpus of spontaneous dialogues in L2 English by French and Japanese L1 speakers for automated assessment of fluency

Résumé

We present a corpus of spontaneous argumentative dialogues in L2 English. The corpus features 10-minute role-plays between pairs of learners from CEFR B1 to C2 level. One hundred French-L1 speakers have been recorded at Grenoble Alpes University, France, and similar data from Japanese-L1 and English-L1 speakers is currently being gathered at Waseda and Doshisha Universities, Japan. An automated processing pipeline, specifically suited for this mutli-speaker spontaneous L2 English, was developed to diarize, transcribe and annotate speech with word and syllable alignment, lexical stress and pause pattern information. The corpus and the pipeline are both open to the academic world.
Fichier principal
Vignette du fichier
CoulangeAl2024_LCSAW_poster.pdf (2.17 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04666131 , version 1 (02-08-2024)

Identifiants

  • HAL Id : hal-04666131 , version 1

Citer

Sylvain Coulange, Takayuki Konishi, Mariko Sugahara, Tsuneo Kato. A corpus of spontaneous dialogues in L2 English by French and Japanese L1 speakers for automated assessment of fluency. 6th International Symposium on Learner Corpus Studies in Asia and the World (LCSAW6), Feb 2024, Kobe, Japan. 2024. ⟨hal-04666131⟩
21 Consultations
5 Téléchargements

Partager

More