TRANSLENDIUM

Introduction

Compromised with the importance of multilinguisme in the global society of technologies of the information and the communication (TIC), Translendium works in the development and improvement of the processing of the natural language and it’s focused on the automatic translation. That is why this is a fundamental and useful tool of communication in the process of globalization we are living. Its target is to develop applications that will enable the translation of texts without human help.

What is the Automatic Translation

Automatic translation between human languages (`Machine Translation’) is a Science Fiction staple, and a long-term scientific dream of enormous social, political, and scientific importance. It was one of the earliest applications suggested for digital computers, but turning this dream into reality has turned out to be a much harder, and in many ways a much more interesting task than at first appeared. Nevertheless, though there remain many outstanding problems, some degree of automatic translation is now a daily reality, and it is likely that during the next decade the bulk of routine technical and business translation will be done with some kind of automatic translation tool, from humble databases containing canned translations of technical terms to genuine Machine Translation Systems that can produce reasonable draft translations (provided the input observes certain restrictions on subject matter, style, and vocabulary).

The automatic translation on line is an area of the computacional linguistic that investigates the use of software to translate texts or speeches of a natural language to another one. In a basic level, the translation for computer allows a simple substitution of the atomic words of a natural language into other language. By means of the use of linguistic bodies it can translate more complicated texts, which allows a more appropriate management of the differences in the linguistic typology, the recognition of phrases, the translation of idiomatic expressions and the isolation of anomalies.

In the last decades, there has been a strong focus in the use of statistical techniques for the development of automatic translation systems. By means of this corpus it estimates parameters of statistical models that establish the probability with which certain words are predisposed to be translated into others, as well as the most probable positions that usually is occupided by the words of the target language depending on the corresponding words of the original sentence.

Second decade of the Automatic Translation (1950-1960)

In 1951 one of the specialists of the prestigious Institute of Technology of Massachusetts (MIT) Yehoshua Bar-Hillel started to work focusing only in TA (Automatic Translation). One year later the first symposium of the TA was organized, with topics as the controlled languages, the sublanguages, the need of the syntax, or the possibility of doing it without the human help. The first public demonstration of an automatic translator was carried out in 1954, in the University of Georgetown, with IBM’s help and the participation of the researcher Leon Dostert. 49 sentences were carefully selected in Russian that were translated into English with a vocabulary of 250 words and 6 grammatical rules. The mediatic success of the demonstration was very good and in the USA they dedicated important amounts of money (the majority given by the Defense Department) to translate Russian, French and German.

It was a moment of initial euphoria, which led to wonder the aim of the fully automatic high quality translation (FAHQT). Between the pioneering developments of that decade it is necessary to emphasize those of the universities of Georgetown and Texas, where there were established the bases of SYSTRAN and METAL.

History

METAL is Siemens-Nixdorf’s system for human-aided machine translation. In the period between 1985 and 1992, a mixed academic and industrial team of on average 5 people (mainly university researchers) has worked on three language pairs of the METAL system: Dutch-French, French-Dutch and French-English. The Leuven researchers have developed extensive computational grammars and lexicons for the languages mentioned. The METAL project has been a good opportunity to acquire a lot of practical know-how in the construction of complex NLP products. Then METAL became Traslendium.

The company Translendium, which is in Barcelona, is in charge of the development of the system of automatic translation Lucy Translator (LT), before that, it was called Comprendium. Translendium is a Catalan subsidiary company of the European group Lucy Software.
The team of Translendium is formed by linguists, lexicographers and engineers with more than 15 years of experience in the field of the automatic translation.

The technology LT is recognized in the world of the automatic translation for the number of language pairs and for the functionality that the system offers.

The system has an engine of translation, with a modular structure of grammatical and lexical computacionals, that allows a morfosintactic analysis of the original text that is transferred later in the desired language. This engine can be connected to translation memory and to a professional lexicon editor.

We can accede to this engine by a distributor servant of multi-user tasks with a web client or with a professional single user client. Different configurations can be created from the product of agreement with the needs of each costumer.

Tools of Translendium

Comprendium Translator Local

Comprendium Translator Local is a translation tool for translators dealing with large amounts of documents and who want to modify the translation parameters.

Functionalities

  • Translation of txt, doc, rtf, html and pdf documents
  • Linguistic translation parameters to control the translation style
  • Colored highlighting of unknown words, alternative translations and compounds
  • Analysis and generation trees
  • Filters to control the translation of specific words
  • Use of translation memories
  • Creation, import and export of translation memories (TMX y TXT)
  • Translation memory module management

Comprendium Translator Net

Comprendium Translator Net is a multi-user translator. Its has been conceived for mid- or large-sized enterprises in which several users need to access the translation services within a network.

The task distribution server enables the creation of large and complex configurations including as many translation engines and final users as necessary. The network configurations can be accessed in two ways: through a web interface or through a PC interface.

Functionalities

  • User Interfaces
    • Translation of text
    • Translation of txt, doc, rtf, html and pdf documents
    • Translation of web pages and all of their links
    • Selection of the subject area
  • Translation server
    • Administration of system parameters
    • Management of language-dependent translation parameters
    • Task load distribution and programming
    • Integration of different types of clients in one same configuration
    • Several translation engines in one configuration
    • Programming APIs

 Lexshop

Lexshop is a tool for dictionary management for professionals. It enables the creation and modification of lexical entries, as well as the import of large volumes of terminology entries. The dictionary manager is the most powerful tool to improve the translation quality.

 Functionalities

  • Access to the complete dictionary architecture
  • Possibility of defining up to 100 lexical features
  • Massive import and export of terminology lists
  • Acceptance of different interchange formats
  • Creation of customer-, product- or domain-specific lexicon modules
  • Intelligent tools to check the lexicon consistency
  • Integration in any configuration

Online Services

The online service allows the immediate document translation, It has 24 language pairs and the quality of the translation depends on the language pair you have chosen. You can freely translate a brief text of a maximum of 4.096 characters and documents’ formats are in txt, doc, rtf and html.

 Protection of Personal Data

All the information that talks about personal data that the user provides across the service Translendium ONLINE is protected by protocol SSL. It means that the information that travels in the Internet from the computer of the costumer up to the servant of Translendium travels it in an encrypted way. The authority of Translendium’s certification ONLINE is Camerfirma. The processing of the documents translated in the service Translendium ONLINE is considered to be confidential. The company promises not to spread your contents and proceed to the elimination of the files once the translation and the checking of the payment and the download is done.

Languages it can translate

Catalan – Spanish

Catalan – English

Catalan – French (prototype)

Spanish – German,

Spanish – English

Spanish – French

English – German

English – French

English – Russian

German – French

German – Russian

English – Italian

Notes

Comment: Translendium, formerly METAL, also Sail-Labs (sail-labs “machine translation” http://www.google.es/search?q=sail-labs+%22machine+translation%22) See also: http://www.ccl.kuleuven.be/about/METAL.html –JosebaAbaitua 12:50, 30 Apr 2008 (CEST)

References

  • Jesús González, A. Giménez, Jorge González, A. L. Lagarda,J. R. Navarro, L. Eliodoro, V. Félix, P. Peris, F. Casacuberta. Una evaluación exhaustiva de Sishitra, un Paradigma Híbrido en TRaducción Automática. (Nov. 2006). Retrieved: 12:46, 19 May 2008 from http://jth2006.unizar.es/finals/4jth_156.pdf

See also

Further Reading

  • Delgado, Ana maría; Cuello, Rafael Oliver; Salomón Sancho, Lourdes (2006): Current Developments in Technology-Assisted Education, Universitat Oberta de Catalunya & Universitat Pompeu Fabra, Spain.(Page 402, 2.2 La traducción automática) Retrieved 12:51, May 12th,2008 from http://www.formatex.org/micte2006/pdf/401-405.pdf
  • González, Jesús; Jiménez A; González, Jorge; Lagarda, A.L; Navarro, J.R; Eliodoro, L; Féliz, V; Peris, P; Casacuberta F (2006): Una evaluación exhaustiva de Sishitra, un paradigma híbrido en traducción automática; IV Jornadas en tecnología del habla, Departamento de sistemas informáticos y computación e Instituto tecnológico de informática, Universidad Politécnica de Valencia, Spain. (Page 96, 5.1 Otros sistemas) Retrieved 13:08, May 12th, 2008 from http://jth2006.unizar.es/finals/4jth_156.pdf
Retrieved from “http://littera.deusto.es/enwiki/index.php/Htl0708:b“

Todavía no hay comentarios »

Tu comentario

HTML-Tags:
<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <pre> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>