Fichier de travail (INPUT) : ./DUMP-TEXT/2-23.txt
Encodage utilisé (INPUT) : utf-8
Forme recherchée : translation|traduction|机器
_________________________________________________________________________________________________
- Ligne n°32 : SYSTRAN - Beyond language - Language translation technologies
Ligne n°68 : ... + Careers- Ligne n°69 : A pioneer and global leader in translation solutions.
Ligne n°70 : Who we are ...- Ligne n°105 : Machine translation (MT) is automated translation. It is the process by
- Ligne n°105 : Machine translation (MT) is automated translation. It is the process by
Ligne n°106 : which computer software is used to translate a text from one natural ...- Ligne n°109 : To process any translation, human or automated, the meaning of a text
Ligne n°110 : in the original (source) language must be fully restored in the target ...
Ligne n°110 : ... in the original (source) language must be fully restored in the target- Ligne n°111 : language, i.e. the translation. While on the surface this seems
Ligne n°112 : straightforward, it is far more complex. Translation is not a mere ...- Ligne n°119 : Human and machine translation each have their share of challenges. For
Ligne n°120 : example, no two individual translators can produce identical ...
Ligne n°120 : ... example, no two individual translators can produce identical- Ligne n°121 : translations of the same text in the same language pair, and it may
Ligne n°122 : take several rounds of revisions to meet customer satisfaction. But the ...
Ligne n°122 : ... take several rounds of revisions to meet customer satisfaction. But the- Ligne n°123 : greater challenge lies in how machine translation can produce
Ligne n°124 : publishable quality translations. ...
Ligne n°123 : ... greater challenge lies in how machine translation can produce- Ligne n°124 : publishable quality translations.
- Ligne n°128 : Rule-based machine translation relies on countless built-in linguistic
Ligne n°129 : rules and millions of bilingual dictionaries for each language pair. ...
Ligne n°138 : ... Translations are built on gigantic dictionaries and sophisticated- Ligne n°139 : linguistic rules. Users can improve the out-of-the-box translation
Ligne n°140 : quality by adding their terminology into the translation process. They ...
Ligne n°139 : ... linguistic rules. Users can improve the out-of-the-box translation- Ligne n°140 : quality by adding their terminology into the translation process. They
Ligne n°141 : create user-defined dictionaries which override the system’s default ...- Ligne n°152 : Statistical machine translation utilizes statistical translation models
- Ligne n°152 : Statistical machine translation utilizes statistical translation models
Ligne n°153 : whose parameters stem from the analysis of monolingual and bilingual ...
Ligne n°153 : ... whose parameters stem from the analysis of monolingual and bilingual- Ligne n°154 : corpora. Building statistical translation models is a quick process,
Ligne n°155 : but the technology relies heavily on existing multilingual corpora. A ...
Ligne n°158 : ... the quality threshold but most companies do not have such large amounts- Ligne n°159 : of existing multilingual corpora to build the necessary translation
Ligne n°160 : models. Additionally, statistical machine translation is CPU intensive ...
Ligne n°159 : ... of existing multilingual corpora to build the necessary translation- Ligne n°160 : models. Additionally, statistical machine translation is CPU intensive
Ligne n°161 : and requires an extensive hardware configuration to run translation ...
Ligne n°160 : ... models. Additionally, statistical machine translation is CPU intensive- Ligne n°161 : and requires an extensive hardware configuration to run translation
Ligne n°162 : models for average performance levels. ...
Ligne n°167 : ... predictable. Dictionary-based customization guarantees improved quality- Ligne n°168 : and compliance with corporate terminology. But translation results may
Ligne n°169 : lack the fluency readers expect. In terms of investment, the ...
Ligne n°173 : ... Statistical MT provides good quality when large and qualified corpora- Ligne n°174 : are available. The translation is fluent, meaning it reads well and
Ligne n°175 : therefore meets user expectations. However, the translation is neither ...
Ligne n°174 : ... are available. The translation is fluent, meaning it reads well and- Ligne n°175 : therefore meets user expectations. However, the translation is neither
Ligne n°176 : predictable nor consistent. Training from good corpora is automated and ...
Ligne n°178 : ... than the specified domain, is poor. Furthermore, statistical MT- Ligne n°179 : requires significant hardware to build and manage large translation
Ligne n°180 : models. ...
Ligne n°181 : ... Rule-Based MT Statistical MT- Ligne n°182 : + Consistent and predictable quality – Unpredictable translation
Ligne n°183 : quality ...
Ligne n°183 : ... quality- Ligne n°184 : + Out-of-domain translation quality – Poor out-of-domain quality
Ligne n°185 : + Knows grammatical rules – Does not know grammar ...
Ligne n°198 : ... Given the overall requirements, there is a clear need for a third- Ligne n°199 : approach through which users would reach better translation quality and
Ligne n°200 : high performance (similar to rule-based MT), with less investment ...
Ligne n°275 : ... SYSTRAN Privacy Policy - SYSTRAN Terms of service - Copyright 2019- Ligne n°276 : SYSTRAN All rights reserved - Online translation software and tools:
Ligne n°277 : Translate Text, Webpage and Files and search Online Dictionary. ...
Ligne n°280 : ... Arabic and Swedish.- Ligne n°281 : Instant messaging translation (MS Lync). www.systranlinks.com