Contextual neural model for translating bilingual multi-speaker conversations

Sameen Maruf; André F.T. Martins; Gholamreza Haffari

Contextual neural model for translating bilingual multi-speaker conversations

Sameen Maruf, André F.T. Martins, Gholamreza Haffari

Department of Data Science & AI

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

17 Citations (Scopus)

Abstract

Recent works in neural machine translation have begun to explore document translation. However, translating online multi-speaker conversations is still an open problem. In this work, we propose the task of translating Bilingual Multi-Speaker Conversations, and explore neural architectures which exploit both source and target-side conversation histories for this task. To initiate an evaluation for this task, we introduce datasets extracted from Europarl v7 and OpenSubtitles2016. Our experiments on four language-pairs confirm the significance of leveraging conversation history, both in terms of BLEU and manual evaluation.

Original language	English
Title of host publication	WMT 2018 - Third Conference on Machine Translation - Proceedings of the Conference
Subtitle of host publication	October 31 - November 1, 2018 Brussels, Belgium
Editors	Ondˇrej Bojar, Rajen Chatterjee, Christian Federmann, Yvette Graham, Barry Haddow, Matthias Huck, Antonio Jimeno Yepes, Philipp Koehn, Christof Monz, Matteo Negri, Aurélie Névéol, Mariana Neves, Matt Post, Lucia Specia, Marco Turchi, Karin Verspoor, Mark Fishel
Place of Publication	Stroudsburg PA USA
Publisher	Association for Computational Linguistics (ACL)
Pages	101-112
Number of pages	12
ISBN (Electronic)	9781948087810
Publication status	Published - 2018
Event	Empirical Methods in Natural Language Processing 2018 - Brussels, Belgium Duration: 31 Oct 2018 → 4 Nov 2018 https://emnlp2018.org/ https://www.aclweb.org/anthology/volumes/D18-1/ (Proceedings)

Conference

Conference	Empirical Methods in Natural Language Processing 2018
Abbreviated title	EMNLP 2018
Country/Territory	Belgium
City	Brussels
Period	31/10/18 → 4/11/18
Internet address	https://emnlp2018.org/ https://www.aclweb.org/anthology/volumes/D18-1/ (Proceedings)

Access to Document

277918264-oaFinal published version, 690 KB

https://www.aclweb.org/anthology/W18-6311

Cite this

Maruf, S., Martins, A. F. T., & Haffari, G. (2018). Contextual neural model for translating bilingual multi-speaker conversations. In O. Bojar, R. Chatterjee, C. Federmann, Y. Graham, B. Haddow, M. Huck, A. Jimeno Yepes, P. Koehn, C. Monz, M. Negri, A. Névéol, M. Neves, M. Post, L. Specia, M. Turchi, K. Verspoor, & M. Fishel (Eds.), WMT 2018 - Third Conference on Machine Translation - Proceedings of the Conference : October 31 - November 1, 2018 Brussels, Belgium (pp. 101-112). Association for Computational Linguistics (ACL). https://www.aclweb.org/anthology/W18-6311

Maruf, Sameen ; Martins, André F.T. ; Haffari, Gholamreza. / Contextual neural model for translating bilingual multi-speaker conversations. WMT 2018 - Third Conference on Machine Translation - Proceedings of the Conference : October 31 - November 1, 2018 Brussels, Belgium. editor / Ondˇrej Bojar ; Rajen Chatterjee ; Christian Federmann ; Yvette Graham ; Barry Haddow ; Matthias Huck ; Antonio Jimeno Yepes ; Philipp Koehn ; Christof Monz ; Matteo Negri ; Aurélie Névéol ; Mariana Neves ; Matt Post ; Lucia Specia ; Marco Turchi ; Karin Verspoor ; Mark Fishel. Stroudsburg PA USA : Association for Computational Linguistics (ACL), 2018. pp. 101-112

@inproceedings{0c47f2ca2c9a4d229e82dffbbf20bba9,

title = "Contextual neural model for translating bilingual multi-speaker conversations",

abstract = "Recent works in neural machine translation have begun to explore document translation. However, translating online multi-speaker conversations is still an open problem. In this work, we propose the task of translating Bilingual Multi-Speaker Conversations, and explore neural architectures which exploit both source and target-side conversation histories for this task. To initiate an evaluation for this task, we introduce datasets extracted from Europarl v7 and OpenSubtitles2016. Our experiments on four language-pairs confirm the significance of leveraging conversation history, both in terms of BLEU and manual evaluation.",

author = "Sameen Maruf and Martins, {Andr{\'e} F.T.} and Gholamreza Haffari",

note = "Funding Information: We thank the reviewers for their useful feedback that helped us in improving our work. We also thank Pierre Lison for providing us with the speaker-annotated OpenSubtitles2016 dataset. This work is supported by the Multi-modal Australian ScienceS Imaging and Visualisation Environment (MASSIVE) (www. massive.org.au), the European Research Council (ERC StG DeepSPIN 758969), the Fundac¸{\~a}o para a Ci{\^e}ncia e Tecnologia through contract UID/EEA/50008/2013, a Google Faculty Research Award to GH and the Australian Research Council through DP160102686. Funding Information: We thank the reviewers for their useful feedback that helped us in improving our work. We also thank Pierre Lison for providing us with the speaker-annotated OpenSubtitles2016 dataset. This work is supported by the Multi-modal Australian ScienceS Imaging and Visualisation Environment (MASSIVE) (www.massive.org.au), the European Research Council (ERC StG DeepSPIN 758969), the Funda??o para a Ci?ncia e Tecnologia through contract UID/EEA/50008/2013, a Google Faculty Research Award to GH and the Australian Research Council through DP160102686. Publisher Copyright: {\textcopyright} 2018 Association for Computational Linguistics.; Empirical Methods in Natural Language Processing 2018, EMNLP 2018 ; Conference date: 31-10-2018 Through 04-11-2018",

year = "2018",

language = "English",

pages = "101--112",

editor = "Ondˇrej Bojar and Chatterjee, {Rajen } and Federmann, {Christian } and Yvette Graham and Barry Haddow and Huck, {Matthias } and {Jimeno Yepes}, Antonio and Koehn, {Philipp } and Christof Monz and Negri, {Matteo } and Aur{\'e}lie N{\'e}v{\'e}ol and Neves, {Mariana } and Post, {Matt } and Specia, {Lucia } and Turchi, {Marco } and Verspoor, {Karin } and Fishel, {Mark }",

booktitle = "WMT 2018 - Third Conference on Machine Translation - Proceedings of the Conference",

publisher = "Association for Computational Linguistics (ACL)",

url = "https://emnlp2018.org/, https://www.aclweb.org/anthology/volumes/D18-1/",

}

Maruf, S, Martins, AFT & Haffari, G 2018, Contextual neural model for translating bilingual multi-speaker conversations. in O Bojar, R Chatterjee, C Federmann, Y Graham, B Haddow, M Huck, A Jimeno Yepes, P Koehn, C Monz, M Negri, A Névéol, M Neves, M Post, L Specia, M Turchi, K Verspoor & M Fishel (eds), WMT 2018 - Third Conference on Machine Translation - Proceedings of the Conference : October 31 - November 1, 2018 Brussels, Belgium. Association for Computational Linguistics (ACL), Stroudsburg PA USA, pp. 101-112, Empirical Methods in Natural Language Processing 2018, Brussels, Belgium, 31/10/18. <https://www.aclweb.org/anthology/W18-6311>

Contextual neural model for translating bilingual multi-speaker conversations. / Maruf, Sameen; Martins, André F.T.; Haffari, Gholamreza.

WMT 2018 - Third Conference on Machine Translation - Proceedings of the Conference : October 31 - November 1, 2018 Brussels, Belgium. ed. / Ondˇrej Bojar; Rajen Chatterjee; Christian Federmann; Yvette Graham; Barry Haddow; Matthias Huck; Antonio Jimeno Yepes; Philipp Koehn; Christof Monz; Matteo Negri; Aurélie Névéol; Mariana Neves; Matt Post; Lucia Specia; Marco Turchi; Karin Verspoor; Mark Fishel. Stroudsburg PA USA : Association for Computational Linguistics (ACL), 2018. p. 101-112.

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - Contextual neural model for translating bilingual multi-speaker conversations

AU - Maruf, Sameen

AU - Martins, André F.T.

AU - Haffari, Gholamreza

N1 - Funding Information: We thank the reviewers for their useful feedback that helped us in improving our work. We also thank Pierre Lison for providing us with the speaker-annotated OpenSubtitles2016 dataset. This work is supported by the Multi-modal Australian ScienceS Imaging and Visualisation Environment (MASSIVE) (www. massive.org.au), the European Research Council (ERC StG DeepSPIN 758969), the Fundac¸ão para a Ciência e Tecnologia through contract UID/EEA/50008/2013, a Google Faculty Research Award to GH and the Australian Research Council through DP160102686. Funding Information: We thank the reviewers for their useful feedback that helped us in improving our work. We also thank Pierre Lison for providing us with the speaker-annotated OpenSubtitles2016 dataset. This work is supported by the Multi-modal Australian ScienceS Imaging and Visualisation Environment (MASSIVE) (www.massive.org.au), the European Research Council (ERC StG DeepSPIN 758969), the Funda??o para a Ci?ncia e Tecnologia through contract UID/EEA/50008/2013, a Google Faculty Research Award to GH and the Australian Research Council through DP160102686. Publisher Copyright: © 2018 Association for Computational Linguistics.

PY - 2018

Y1 - 2018

N2 - Recent works in neural machine translation have begun to explore document translation. However, translating online multi-speaker conversations is still an open problem. In this work, we propose the task of translating Bilingual Multi-Speaker Conversations, and explore neural architectures which exploit both source and target-side conversation histories for this task. To initiate an evaluation for this task, we introduce datasets extracted from Europarl v7 and OpenSubtitles2016. Our experiments on four language-pairs confirm the significance of leveraging conversation history, both in terms of BLEU and manual evaluation.

AB - Recent works in neural machine translation have begun to explore document translation. However, translating online multi-speaker conversations is still an open problem. In this work, we propose the task of translating Bilingual Multi-Speaker Conversations, and explore neural architectures which exploit both source and target-side conversation histories for this task. To initiate an evaluation for this task, we introduce datasets extracted from Europarl v7 and OpenSubtitles2016. Our experiments on four language-pairs confirm the significance of leveraging conversation history, both in terms of BLEU and manual evaluation.

UR - http://www.scopus.com/inward/record.url?scp=85115105690&partnerID=8YFLogxK

M3 - Conference Paper

AN - SCOPUS:85115105690

SP - 101

EP - 112

BT - WMT 2018 - Third Conference on Machine Translation - Proceedings of the Conference

A2 - Bojar, Ondˇrej

A2 - Chatterjee, Rajen

A2 - Federmann, Christian

A2 - Graham, Yvette

A2 - Haddow, Barry

A2 - Huck, Matthias

A2 - Jimeno Yepes, Antonio

A2 - Koehn, Philipp

A2 - Monz, Christof

A2 - Negri, Matteo

A2 - Névéol, Aurélie

A2 - Neves, Mariana

A2 - Post, Matt

A2 - Specia, Lucia

A2 - Turchi, Marco

A2 - Verspoor, Karin

A2 - Fishel, Mark

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg PA USA

T2 - Empirical Methods in Natural Language Processing 2018

Y2 - 31 October 2018 through 4 November 2018

ER -

Maruf S, Martins AFT, Haffari G. Contextual neural model for translating bilingual multi-speaker conversations. In Bojar O, Chatterjee R, Federmann C, Graham Y, Haddow B, Huck M, Jimeno Yepes A, Koehn P, Monz C, Negri M, Névéol A, Neves M, Post M, Specia L, Turchi M, Verspoor K, Fishel M, editors, WMT 2018 - Third Conference on Machine Translation - Proceedings of the Conference : October 31 - November 1, 2018 Brussels, Belgium. Stroudsburg PA USA: Association for Computational Linguistics (ACL). 2018. p. 101-112

Contextual neural model for translating bilingual multi-speaker conversations

Abstract

Conference

Access to Document

Other files and links

Learning Deep Semantics for Automatic Translation between Human Languages

Cite this

Contextual neural model for translating bilingual multi-speaker conversations

Abstract

Conference

Access to Document

Other files and links

Projects

Learning Deep Semantics for Automatic Translation between Human Languages

Cite this