Spoken Language Translation Graphs Re-decoding using Automatic Quality Assessment

Laurent Besacier; Benjamin Lecouteux; Ngoc Luong; Ngoc Le

Poster communications

Spoken Language Translation Graphs Re-decoding using Automatic Quality Assessment

Laurent Besacier ^{1, 2} Benjamin Lecouteux ^{1, 2, 3} Ngoc Luong ^{2, 1} Ngoc Le ^{2, 1}

1 LIG - Laboratoire d'Informatique de Grenoble

2 GETALP - Groupe d’Étude en Traduction Automatique/Traitement Automatisé des Langues et de la Parole

LIG - Laboratoire d'Informatique de Grenoble

3 UGA [2016-2019] - Université Grenoble Alpes [2016-2019]

Abstract : This paper investigates how automatic quality assessment of spoken language translation (SLT) can help re-decoding SLT output graphs and improving the overall speech translation performance. Using robust word confidence measures (from both ASR and MT) to re-decode the SLT graph leads to a significant BLEU improvement (more than 2 points) compared to our SLT baseline (French-English task).

Document type :

Poster communications

Domain :

Computer Science [cs] / Computation and Language [cs.CL]

Complete list of metadata

Cited literature [5 references] Display Hide Download

https://hal.archives-ouvertes.fr/hal-02095256
Contributor : Benjamin Lecouteux Connect in order to contact the contributor
Submitted on : Wednesday, April 10, 2019 - 12:11:20 PM
Last modification on : Thursday, October 21, 2021 - 3:47:02 AM

Spoken Language Translation Graphs Re-decoding using Automatic Quality Assessment

File

Identifiers

Collections

Citation

Export

Metrics

26

36

Spoken Language Translation Graphs Re-decoding using Automatic Quality Assessment

File

Identifiers

Collections

Citation

Export

Share

Metrics

26

36