SYSTRAN Purely Neural MT Engines for WMT2017
نویسندگان
چکیده
This paper describes SYSTRAN’s systems submitted to the WMT 2017 shared news translation task for English-German, in both translation directions. Our systems are built using OpenNMT1, an opensource neural machine translation system, implementing sequence-to-sequence models with LSTM encoder/decoders and attention. We experimented using monolingual data automatically back-translated. Our resulting models are further hyperspecialised with an adaptation technique that finely tunes models according to the evaluation test sentences.
منابع مشابه
How Portable are Controlled Languages Rules? A Comparison of Two Empirical MT Studies
This paper describes two studies on the effectiveness of Controlled Language (CL) rules for MT. Both studies investigated the language pair English-German and used corpora from the IT domain. However, they differ in terms of the MT engines employed (Systran vs. IBM WebSphere) and the evaluative methodologies used. Study A examines the effectiveness of CL rules by measuring temporal, technical a...
متن کاملSystran Mt Dictionary Development
SYSTRAN has demonstrated success in the MT field with its long history spanning nearly 30 years. As a general-purpose fully automatic MT system, SYSTRAN employs a transfer approach. Among its several components, large, carefully encoded, high-quality dictionaries are critical to SYSTRAN's translation capability. A total of over 2.4 million words and expressions are now encoded in the dictionari...
متن کاملContent Management and Retrieval Software Bulletin SYSTRAN and the Reinvention of MT
Yes. With its newly architected system, its persuasive hold on the multilingual Web-browsing market, and its successful incursions into the customer service area, SYSTRAN is poised to achieve a level of success that has not yet been realized by any other MT company and perhaps to redefine the field of uses for automatic translation solutions. SYSTRAN will need to move cautiously, given MT’s his...
متن کاملSpeech and Natural Language Software Bulletin Systran Enables Multilingual Customer Support for Autodesk
Autodesk’s deployment of Systran’s machine translation (MT) technology is an early example of enterprise machine translation for multilingual customer support. IDC believes the implementation of MT by a company of Autodesk’s scale and global reach represents the beginning of new commercial opportunities for Systran and other MT vendors. The return on investment (ROI) for applications that subst...
متن کاملStone Soup and the French Room
The paper argues that the IBM statistical approach to machine translation has done rather better after a few years than many sceptics believed it could. However, it is neither as novel as its proponents suggest nor is it making claims as clear and simple as they would have us believe. The performance of the purely statistical system (and we discuss what that phrase could mean) has not equaled t...
متن کامل