A roadmap for MT : four « keys » to handle more languages, for all kinds of tasks, while making it possible to improve quality (on demand)

Despite considerable investment over the past 50 years, only a small number of language pairs is covered by MT systems designed for information access, and even fewer are capable of quality translation or speech translation. To open the door toward MT of adequate quality for all languages (at least in principle), we propose four keys. On the technical side, we should (1) dramatically increase the use of learning techniques which have demonstrated their potential at the research level, and (2) use pivot architectures, the most universally usable pivot being UNL. On the organizational side, the keys are (3) the cooperative development of open source linguistic resources on the Web, and (4) the construction of systems where quality can be improved "on demand" by users, either a priori through interactive disambiguation, or a posteriori by correcting the pivot representation through any language, thereby unifying MT, computer-aided authoring, and multilingual generation.

[1]  Alon Lavie,et al.  An interlingua based on domain actions for machine translation of task-oriented dialogues , 1998, ICSLP.

[2]  Wolfgang Wahlster,et al.  Verbmobil: Foundations of Speech-to-Speech Translation , 2000, Artificial Intelligence.

[3]  Christian Boitet,et al.  Multilingual Dialogue-Based MT for monolingual authors: the LIDIA project and a first mockup , 1994, Machine Translation.

[4]  José Coch,et al.  Interactive Multilingual Generation , 2001, CICLing.

[5]  Christian BOITET,et al.  A way to integrate context processing in the MT component of spoken, task-oriented translation systems , 2000 .

[6]  Eric Wehrli,et al.  The Ips System , 1992, COLING.

[7]  Muriel Vasconcellos,et al.  SPANAM and ENGSPAN: Machine Translation at the Pan American Health Organization , 1985, Comput. Linguistics.

[8]  Christian Boitet,et al.  On UNL as the future “html of the linguistic content” & the reuse of existing NLP components in UNL-related applications with the example of a UNL-French deconverter , 2000, COLING.

[9]  John Lehrberger,et al.  Machine Translation: Linguistic characteristics of MT systems and general methodology of evaluation , 1988 .

[10]  Eiichiro Sumita,et al.  Solutions to Problems Inherent in Spoken-language Translation: The ATR-MATRIX Approach , 1999 .

[11]  Christian Boitet,et al.  Analysis into a formal task-oriented pivot without clear abstract - semantics is best handled as "usual" translation , 2000, INTERSPEECH.

[12]  Sergei Nirenburg,et al.  Knowledge-based machine translation , 1989, COLING.

[13]  Teruko Mitamura,et al.  The KANT System: Fast, Accurate, High-Quality Translation in Practical Domains , 1992, COLING.

[14]  Toshiyuki Takezawa,et al.  End-to-end evaluation in ATR-MATRIX: speech translation system between English and Japanese , 1999, EUROSPEECH.

[15]  M. Nagao,et al.  Machine Translation Summit , 1989 .