Evaluating Subtitle Segmentation for End-to-end Generation Systems

Autor: Alina Karakanta, François Buet, Mauro Cettolo, François Yvon

Publikováno v: Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022 )
13th Language Resources and Evaluation Conference (LREC)
13th Language Resources and Evaluation Conference (LREC), ELDA, Jun 2022, Marseille, France
HAL

Subtitles appear on screen as short pieces of text, segmented based on formal constraints (length) and syntactic/semantic criteria. Subtitle segmentation can be evaluated with sequence segmentation metrics against a human reference. However, standard

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9737d0660d6c5e11ac9282da848a1090

CTC-based Compression for Direct Speech Translation

Autor: Mauro Cettolo, Matteo Negri, Marco Gaido, Marco Turchi

Publikováno v: Università degli di Trento-IRIS
Scopus-Elsevier
EACL

Previous studies demonstrated that a dynamic phone-informed compression of the input audio is beneficial for speech translation (ST). However, they required a dedicated model for phone recognition and did not test this solution for direct ST, in whic

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::5649981242a11a730e9599d9fb1ab161
http://arxiv.org/abs/2102.01578

Contextualized Translation of Automatically Segmented Speech

Autor: Marco Gaido, Marco Turchi, Mattia Antonino Di Gangi, Matteo Negri, Mauro Cettolo

Publikováno v: INTERSPEECH

Direct speech-to-text translation (ST) models are usually trained on corpora segmented at sentence level, but at inference time they are commonly fed with audio split by a voice activity detector (VAD). Since VAD segmentation is not syntax-informed,

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::73e9500bb65a3a7db12651eced707009
http://arxiv.org/abs/2008.02270

Neural versus phrase-based MT quality: An in-depth analysis on English-German and English-French

Autor: Mauro Cettolo, Marcello Federico, Arianna Bisazza, Luisa Bentivogli

Publikováno v: Computer Speech and Language, 49, 52-70. Elsevier

Within the field of statistical machine translation, the neural approach (NMT) is currently pushing ahead the state of the art performance traditionally achieved by phrase-based approaches (PBMT), and is rapidly becoming the dominant technology in ma

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::088d7b127b916adf161cde610ec885e3
https://hdl.handle.net/1887/80980

Pronoun-Focused MT and Cross-Lingual Pronoun Prediction: Findings of the 2015 DiscoMT Shared Task on Pronoun Translation

Autor: Jörg Tiedemann, Mauro Cettolo, Sara Stymne, Yannick Versley, Christian Hardmeier, Preslav Nakov

Publikováno v: DiscoMT@EMNLP
Hardmeier, C, Nakov, P, Stymne, S, Tiedemann, J, Versley, Y & Cettolo, M 2015, Pronoun-Focused MT and Cross-Lingual Pronoun Prediction: Findings of the 2015 DiscoMT Shared Task on Pronoun Translation . in Proceedings of the Second Workshop on Discourse in Machine Translation . Lisbon, Portugal, pp. 1-16, Second Workshop on Discourse in Machine Translation, Lisbon, Portugal, 17/09/15 . https://doi.org/10.18653/v1/W15-2501

We describe the design, the evaluation setup, and the results of the DiscoMT 2015 shared task, which included two sub-tasks, relevant to both the machine translation (MT) and the discourse communities: (i) pronoun-focused translation, a practical MT

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::dd88213eb4cfb690136f85386412d8e2
http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-262364