Guardado en:
Detalles Bibliográficos
Autores principales: Sperber, Matthias, Bojar, Ondřej, Haddow, Barry, Javorský, Dávid, Ma, Xutai, Negri, Matteo, Niehues, Jan, Polák, Peter, Salesky, Elizabeth, Sudoh, Katsuhito, Turchi, Marco
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:https://arxiv.org/abs/2406.03881
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866917686495150080
author Sperber, Matthias
Bojar, Ondřej
Haddow, Barry
Javorský, Dávid
Ma, Xutai
Negri, Matteo
Niehues, Jan
Polák, Peter
Salesky, Elizabeth
Sudoh, Katsuhito
Turchi, Marco
author_facet Sperber, Matthias
Bojar, Ondřej
Haddow, Barry
Javorský, Dávid
Ma, Xutai
Negri, Matteo
Niehues, Jan
Polák, Peter
Salesky, Elizabeth
Sudoh, Katsuhito
Turchi, Marco
contents Human evaluation is a critical component in machine translation system development and has received much attention in text translation research. However, little prior work exists on the topic of human evaluation for speech translation, which adds additional challenges such as noisy data and segmentation mismatches. We take first steps to fill this gap by conducting a comprehensive human evaluation of the results of several shared tasks from the last International Workshop on Spoken Language Translation (IWSLT 2023). We propose an effective evaluation strategy based on automatic resegmentation and direct assessment with segment context. Our analysis revealed that: 1) the proposed evaluation strategy is robust and scores well-correlated with other types of human judgements; 2) automatic metrics are usually, but not always, well-correlated with direct assessment scores; and 3) COMET as a slightly stronger automatic metric than chrF, despite the segmentation noise introduced by the resegmentation step systems. We release the collected human-annotated data in order to encourage further investigation.
format Preprint
id arxiv_https___arxiv_org_abs_2406_03881
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Evaluating the IWSLT2023 Speech Translation Tasks: Human Annotations, Automatic Metrics, and Segmentation
Sperber, Matthias
Bojar, Ondřej
Haddow, Barry
Javorský, Dávid
Ma, Xutai
Negri, Matteo
Niehues, Jan
Polák, Peter
Salesky, Elizabeth
Sudoh, Katsuhito
Turchi, Marco
Computation and Language
Human evaluation is a critical component in machine translation system development and has received much attention in text translation research. However, little prior work exists on the topic of human evaluation for speech translation, which adds additional challenges such as noisy data and segmentation mismatches. We take first steps to fill this gap by conducting a comprehensive human evaluation of the results of several shared tasks from the last International Workshop on Spoken Language Translation (IWSLT 2023). We propose an effective evaluation strategy based on automatic resegmentation and direct assessment with segment context. Our analysis revealed that: 1) the proposed evaluation strategy is robust and scores well-correlated with other types of human judgements; 2) automatic metrics are usually, but not always, well-correlated with direct assessment scores; and 3) COMET as a slightly stronger automatic metric than chrF, despite the segmentation noise introduced by the resegmentation step systems. We release the collected human-annotated data in order to encourage further investigation.
title Evaluating the IWSLT2023 Speech Translation Tasks: Human Annotations, Automatic Metrics, and Segmentation
topic Computation and Language
url https://arxiv.org/abs/2406.03881