MMT will deliver a language independent commercial online translation service based on a new open-source machine translation distributed architecture

Informations projet

MMT

N° de convention de subvention: 645487

DOI

10.3030/645487

Projet clôturé

Date de signature de la CE 12 Decembre 2014

Date de début 1 Janvier 2015

Date de fin 31 Decembre 2017

Financé au titre de

• INDUSTRIAL LEADERSHIP - Leadership in enabling and industrial technologies - Information and Communication Technologies (ICT)

Coût total

€ 3 695 200,00

Contribution de l’UE

€ 2 994 700,00

Coordonné par

TRANSLATED SRL
Italy

CORDIS fournit des liens vers les livrables publics et les publications des projets HORIZON.

Les liens vers les livrables et les publications des projets du 7e PC, ainsi que les liens vers certains types de résultats spécifiques tels que les jeux de données et les logiciels, sont récupérés dynamiquement sur OpenAIRE .

Livrables

Other (3)

Open source distribution

Release of the source code and relative documentation.

Live Integration with TAUS and MyMemory

Integration of MMT with existing technologies from TAUS and Translated to expose the MMT technology to million of users.

Open Source Installer, Billion words system with APIs & live interface

Release of an open source installer providing API to connect to third party services and a web interface.

Documents, reports (14)

Legal framework for the usage of data collected

Reports on the most suitable legal framework under which to release the index of parallel data collected.

Third Report on Database and MT Infrastructure

Reports on the implementation of the databases, models, and decoder.

Data Management Plan

Document describing how data will be managed during the project which data will be collected generated and published as a result of the activities of the project It will describe how the data will be exploited and why some data cannot be shared Information on storing and maintaining the data will also be provided

Second Evaluation Plan Report

Revises the original evaluation plan.

Second Report on Data Supply

Reports the status of the data repository.

Report on Data Repository

Describes the structure of the data repository and the software to collect and align parallel data from the web.

First Design and Specifications Report

Reports the preliminary design of the whole infrastructure and its components required to their actual implementation done in WP3

First Report on Data Supply

Reports the status of the data repository.

First Evaluation Plan Report

Reports on the evaluation plan

Second Technology Assessment Report

Reports on the MT components, performance and scalability testing.

First Report on Database and MT Infrastructure

Reports on the implementation of the databases, models, and decoder.

Second Report on Database and MT Infrastructure

Reports on the implementation of the databases, models, and decoder.

Second Design and Specifications Report

Revises and finalizes the initial design provided by D1.1 taken into account the feedback from WP3 and WP5.

First Technology Assessment Report

Reports on the MT components, performance and scalability testing.

Publications

Conference proceedings (27)

Neural vs. Phrase-Based Machine Translation in a Multi-Domain Scenario

Auteurs: M. Amin Farajian, Marco Turchi, Matteo Negri, Nicola Bertoldi, and Marcello Federico
Publié dans: Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 2, Short Papers, 2017
Éditeur: European Chapter of the Association for Computational Linguistics

MMT: New Open Source MT for the Translation Industry

Auteurs: Nicola Bertoldi, Roldano Cattoni, Mauro Cettolo, Amin Farajian, Marcello Federico, Davide Caroselli, Luca Mastrostefano, Andrea Rossi, Marco Trombetti, Ulrich Germann, and David Madl
Publié dans: Proceedings of the 20th Annual Conference of the European Association for Machine Translation, 2017
Éditeur: European Association for Machine Translation

Guiding Neural Machine Translation Decoding with External Knowledge

Auteurs: Rajen Chatterjee, Matteo Negri, Marco Turchi, Marcello Federico, Lucia Specia, and Frédéric Blain
Publié dans: Proceedings of the 2nd Conference on Machine Translation (WMT17), Volume 1: Research Papers, 2017
Éditeur: WMT

FBK’s Participation to the English-to-German News Translation Task of WMT 2017

Auteurs: Mattia A. Di Gangi, Nicola Bertoldi and Marcello Federico
Publié dans: Proceedings of the 2nd Conference on Machine Translation (WMT17), Volume 2: Shared Task Papers, 2017
Éditeur: WMT

Multi-Domain Neural Machine Translation through Unsupervised Adaptation

Auteurs: M. A. Farajian, Marco Turchi, Matteo Negri and Marcello Federico
Publié dans: Proceedings of the 2nd Conference on Machine Translation (WMT17), Volume 1: Research Papers, 2017
Éditeur: WMT

Can Monolingual Embeddings Improve Neural Machine Translation?

Auteurs: Mattia Antonino Di Gangi, and Marcello Federico
Publié dans: Proceedings of the 4th Italian Conference on Computational Linguistics (CLIC-IT), 2017
Éditeur: CLIC-IT

Multilingual Neural Machine Translation for Low Resource Languages

Auteurs: Surafel Melaku Lakew, Mattia Antonino Di Gangi, and Marcello Federico
Publié dans: Proceedings of the 4th Italian Conference on Computational Linguistics (CLIC-IT), 2017
Éditeur: CLIC-IT

Monolingual Embeddings for Low Resourced Neural Machine Translation

Auteurs: Mattia Antonino Di Gangi and Marcello Federico
Publié dans: Proceedings of the 14th International Workshop on Spoken Language Translation, 2017
Éditeur: IWSLT

Improving Zero-Shot Translation of Low-Resource Languages

Auteurs: Surafel Melaku Lakew, Quintino Francesco Lotito, Marco Turchi, Matteo Negri, and Marcello Federico
Publié dans: Proceedings of the 14th International Workshop on Spoken Language Translation, 2017
Éditeur: IWSLT

FBK’s Multilingual Neural Machine Translation System for IWSLT 2017

Auteurs: Surafel Melaku Lakew, Quintino Francesco Lotito, Marco Turchi, Matteo Negri, and Marcello Federico
Publié dans: Proceedings of the 14th International Workshop on Spoken Language Translation, 2017
Éditeur: IWSLT

The University of Edinburgh's Neural MT systems for WMT17

Auteurs: Rico Sennrich, Alexandra Birch, Anna Currey, Ulrich Germann, Barry Haddow, Kenneth Heafield, Antonio Valerio Miceli Barone, and Philip Williams.
Publié dans: Proceedings of the Second Conference on Machine Translation, 2017
Éditeur: ACL

An Exploration of Neural Sequence-to-Sequence Architectures for Automatic Post-Editing

Auteurs: Marcin Junczys-Dowmunt, Roman Grundkiewicz
Publié dans: Proceedings of the Eighth International Joint Conference on Natural Language Processing (Volume 1: Long Papers), 2017
Éditeur: Asian Federation of Natural Language Processing

Regularization techniques for fine-tuning in neural machine translation

Auteurs: Antonio Valerio Miceli Barone, Barry Haddow, Ulrich Germann, and Rico Sennrich
Publié dans: Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, 2017
Éditeur: ACL

WAGS: A Beautiful English-Italian Benchmark Supporting Word Alignment Evaluation of Rare Words

Auteurs: L. Bentivogli, M. Cettolo, M. A. Farajian, M. Federico
Publié dans: Proceedings of the 10th Language Resources and Evaluation Conference, 2016, Page(s) 3535-3542
Éditeur: European Language Resources Association

TMOP: A Tool for Unsupervised Translation Memory Cleaning

Auteurs: M. J. Sabet, M. Negri, M. Turchi, J. G. C. de Souza, M. Federico
Publié dans: Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, Numéro System Demonstrations, August 7-12, 2016, 2016, Page(s) 49-54
Éditeur: Association for Computational Linguistics

An Unsupervised Method for Automatic Translation Memory Cleaning

Auteurs: M. J. Sabet, M. Negri, Marco Turchi, E. Barbu
Publié dans: Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, Numéro August 7-12, 2016, 2016, Page(s) 287–292
Éditeur: Association for Computational Linguistics

MT Adaptation from TMs in ModernMT

Auteurs: M. Federico
Publié dans: Proceedings of 12th Conference of the Association for Machine Translation in the Americas, Numéro vol. 2: MT Users' Track, 2016, Page(s) 19-57
Éditeur: Association for Machine Translation in the Americas

Neural versus Phrase-Based Machine Translation Quality: a Case Study

Auteurs: L. Bentivogli, A. Bisazza, M. Cettolo, M. Federico
Publié dans: Proceedings of Conference on Empirical Methods in Natural Language Processing, Numéro November 1-5, 2016, 2016, Page(s) 257-267
Éditeur: Association for Computational Linguistics

FBK's Neural Machine Translation Systems for IWSLT 2016

Auteurs: M. A. Farajian, R. Chatterjee, C. Conforti, S. Jalalvand, V. Balaraman, M. A. Di Gangi, D. Ataman, M. Turchi, M. Negri, M. Federico,
Publié dans: Proceedings of the 13th Workshop on Spoken Language Translation, Numéro December 8-9, 2016, 2016
Éditeur: IWSLT

The IWSLT 2016 Evaluation Campaign

Auteurs: M. Cettolo, J. Niehues, S. Stüker, L. Bentivogli, M. Federico
Publié dans: Proceedings of the 13th Workshop on Spoken Language Translation, Numéro December 8-9, 2016, 2016
Éditeur: IWSLT

Bilingual Document Alignment with Latent Semantic Indexing

Auteurs: Ulrich Germann
Publié dans: Proceedings of the First Conference on Machine Translation, Numéro Vol. 2, Page(s) 692–696
Éditeur: Association for Computational Linguistics

N-gram Language Models for Massively Parallel Devices

Auteurs: Nikolay Bogoychev and Adam Lopez
Publié dans: Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics, 2016
Éditeur: Association for Computational Linguistics

Fast and Highly Parallelizable Phrase Table for Statistical Machine Translation

Auteurs: Nikolay Bogoychev and Hieu Hoang
Publié dans: Proceedings of the First Conference on Machine Translation, Numéro Vol. 1, 2016, Page(s) 102–109
Éditeur: Association for Computational Linguistics

The Reasonable Effectiveness of Data

Auteurs: A. Ruopp
Publié dans: Proceedings of 12th Conference of the Association for Machine Translation in the Americas, 2016, Page(s) 123-142
Éditeur: Association for Machine Translation in the Americas

Findings of the WMT 2016 Bilingual Document Alignment Shared Task

Auteurs: C. Buck, P. Koehn
Publié dans: Proceedings of the First Conference on Machine Translation, Numéro Volume 2: Shared Task Papers , August 11-12, 2016, 2016, Page(s) 554–563
Éditeur: Association for Computational Linguistics

Topic Adaptation for Machine Translation of E-Commerce Content

Auteurs: P. Mathur, M. Federico, S. Köprü, S. Khadivi, H. Sawaf
Publié dans: Proceedings of MT Summit XV, Numéro vol. 1: MT Researchers' Track, 30 October-3 November, 2015, 2015, Page(s) 270-281
Éditeur: Association for Machine Translation in the Americas

MT Quality Estimation for E-Commerce Data

Auteurs: J. G. C. de Souza, M. Federico, H. Sawaf
Publié dans: Proceedings of MT Summit XV, Numéro vol. 2: MT Users' Track, 30 October - 3 November, 2015, 2015, Page(s) 20-29
Éditeur: Association for Machine Translation in the Americas

Peer reviewed articles (2)

Automatic translation memory cleaning

Auteurs: Matteo Negri, Duygu Ataman, Masoud Jalili Sabet, Marco Turchi, Marcello Federico
Publié dans: Machine Translation, Numéro 31/3, 2017, Page(s) 93-115, ISSN 0922-6567
Éditeur: Kluwer Academic Publishers
DOI: 10.1007/s10590-017-9191-5

Sampling Phrase Tables for the Moses Statistical Machine Translation System

Auteurs: Ulrich Germann
Publié dans: The Prague Bulletin of Mathematical Linguistics, Numéro No. 104, October 2015, 2015, Page(s) 39-50, ISSN 0032-6585
Éditeur: The Prague Bulletin of Mathematical Linguistics
DOI: 10.1515/pralin-2015-0012

Other (1)

Modern MT: A New Open-Source Machine Translation Platform for the Translation Industry

Auteurs: U. Germann, E. Barbu, L. Bentivogli, N. Bertoldi, N. Bogoychev, C. Buck, D. Caroselli, L. Carvalho, A. Cattelan, R. Cattoni, M. Cettolo, M. Federico, B. Haddow, D. Madl, L. Mastrostefano, P. Mathur, A. Ruopp, A. Samiotou, V. Sudharshan, M. Trombetti, J. van der Meer
Publié dans: Baltic Journal of Modern Computing, Numéro Vol. 4, No. 2, 2016, Page(s) 397–397
Éditeur: University of Latvia