从计算机辅助翻译到协同翻译

由于机器翻译系统的译文质量仍难以达到实用化要求,计算机辅助翻译技术逐渐成为研究热点,并且取得了很好的实际效果,大大提高了翻译产业的生产率。随着辅助翻译规模的不断扩大,多名在空间上分散的用户被组织起来共同完成一项翻译任务已成为普遍现象,这种新的翻译模式称为协同翻译。该文对计算机辅助翻译和协同翻译技术进行综述,首先从辅助译文生成、译后编辑和系统反馈学习等方面介绍了计算机辅助翻译技术的常用方法和研究进展,随后讨论了计算机辅助翻译与协同翻译之间的联系和区别,分析了协同翻译技术所面临的主要问题,并介绍了现有研究的解决方法。最后对协同翻译的未来发展方向进行了展望。     

Maximum Entropy Modeling: A Suitable Framework to Learn Context-Dependent Lexicon Models for Statistical Machine Translation

Current statistical machine translation systems are mainly based on statistical word lexicons. However, these models are usually context-independent, therefore, the disambiguation of the translation of a source word must be carried out using other probabilistic distributions (distortion distributions and statistical language models). One efficient way to add contextual information to the statistical lexicons is based on maximum entropy modeling. In that framework, the context is introduced through feature functions that allow us to automatically learn context-dependent lexicon models.In a first approach, maximum entropy modeling is carried out after a process of learning standard statistical models (alignment and lexicon). In a second approach, the maximum entropy modeling is integrated in the expectation-maximization process of learning standard statistical models.Experimental results were obtained for two well-known tasks, the French–English Canadian Parliament Hansards task and the German–English Verbmobil task. These results proved that the use of maximum entropy models in both approaches, can help to improve the performance of the statistical translation systems.This work has been partially supported by the European Union under grant IST-2001-32091 and by the Spanish CICYT under project TIC-2003-08681-C02-02. The experiments on the Verbmobil task were done when the first author was a visiting scientist at RWTH Aachen-Germany.Editors: Dan Roth and Pascale Fung     

Machine assisted translation from English to a Slavic language

The paper discusses a methodology of machine assisted translation (MAT) as an alternative to fully automatic MT. A prototype MAT system is described, which is an integration of a dictionary database system and a text editor. The functional requests for such a system from the linguistic point of view and some general problems of designing and implementing such systems are presented. Special attention is given to language-dependence, and to the problem of completeness and efficiency of the linguistic data representation for a very simple system. A statistic analysis of English inflexion and word derivation patterns is presented.     

基于Lucene的英汉跨语言信息检索

描述了一个英汉跨语言检索系统的设计与实现,其主要研究目的在于寻找更为有效的英汉查询翻译方法,以及提高汉语检索系统的性能。在英汉查询翻译方面,以英汉双语词典为基础,建立了查询翻译算法。在汉语检索方面,分析不同索引单元对于检索性能的影响,基于Lucene全文索引工具包建立了搜索引擎。在系统评测方面,提出了一种根据主题,快速构建评测数据的方法。     

机器翻译系统及翻译工具的开发和使用

本文综述了当前计算机翻译软件的需求和应用情况,着重讨论了如何设计翻译系统使其质量能达到可出版水平, 其中包括开发受限语言翻译系统, 翻译工作站, 及软件本地化。同时本文还涉及到如何开发非传统概念的翻译软件,特别是针对Web页面和其它应用在Internet上各类信息的翻译软件。本文还讨论了机器翻译未来的需要以及正在开发的一些系统。文中最后一部分比较了一下通过人译、机译,机助人译的方式达到的几种最合适的翻译方法。     

The role and design of computer studies in a research-oriented translation program

With increasing human-machine interaction in the professional translator's work environment, more and more translator training programs are launching translation-specific computer studies. This paper focuses on the research-oriented, as opposed to the practically-oriented, translation program. We argue that computer studies in such a program should prepare students for research at either the receiving or production ends of machine translation systems, both of which require linguistic, computational and translational expertise. We discuss some general considerations for the design of such computer studies, based on a seminar given in the M.A. Translation program at the University of Ottawa, Canada.Ingrid Meyer is an assistant professor in the School of Translators and Interpreters, University of Ottawa. Her research interests are machine translation and computational lexicography.     

机器辅助翻译中模糊查词典和快速录入单词

本文介绍在单词记忆不准确的情况下, 如何查找词典以及如何只键入单词中的几个字母快速录入单词的算法。在辅助翻译和写作系统中, 词汇级的帮助是最基本的, 主要指词典查询。但很多情况下, 用户单词记忆不很准确, 只记住了几个字母, 本文解决这种情况下的模糊查询问题。这种模糊技术的核心是全文检索, 依赖于词典的特殊索引。在解决了模糊查询之后, 利用全文检索技术以及模糊二分查找技术进一步开发了写作系统中的快速录入功能。     

稀缺资源语言神经网络机器翻译研究综述

作为目前主流翻译方法的神经网络机器翻译已经取得了很大突破, 在很多具有丰富数据资源的语言上的翻译质量也不断得到改善, 但对于稀缺资源语言的翻译效果却仍然并不理想. 稀缺资源语言机器翻译是目前机器翻译领域的重要研究热点之一, 近几年来吸引了国内外的广泛关注. 本文对稀缺资源语言机器翻译的研究进行比较全面的回顾, 首先简要介绍了与稀缺资源语言翻译相关的学术活动和数据集, 然后重点梳理了目前主要的研究方法和一些研究结论, 总结了每类方法的特点, 在此基础上总结了不同方法之间的关系并分析了目前的研究现状. 最后, 对稀缺资源语言机器翻译未来可能的研究趋势和发展方向进行了展望,并给出了相关建议.     

Automatic extraction of translations from web-based bilingual materials

This paper describes the framework of the StatCan Daily Translation Extraction System (SDTES), a computer system that maps and compares web-based translation texts of Statistics Canada (StatCan) news releases in the StatCan publication The Daily. The goal is to extract translations for translation memory systems, for translation terminology building, for cross-language information retrieval and for corpus-based machine translation systems. Three years of officially published statistical news release texts at were collected to compose the StatCan Daily data bank. The English and French texts in this collection were roughly aligned using the Gale-Church statistical algorithm. After this, boundary markers of text segments and paragraphs were adjusted and the Gale-Church algorithm was run a second time for a more fine-grained text segment alignment. To detect misaligned areas of texts and to prevent mismatched translation pairs from being selected, key textual and structural properties of the mapped texts were automatically identified and used as anchoring features for comparison and misalignment detection. The proposed method has been tested with web-based bilingual materials from five other Canadian government websites. Results show that the SDTES model is very efficient in extracting translations from published government texts, and very accurate in identifying mismatched translations. With parameters tuned, the text-mapping part can be used to align corpus data collected from official government websites; and the text-comparing component can be applied in prepublication translation quality control and in evaluating the results of statistical machine translation systems.     

ΦDmDialog: A speech-to-speech dialogue translation system

Development of interpreting telephony or a speech-to-speech translation system is one of the ultimate goals of speech recognition, natural language processing, artificial intelligence and machine translation. This paper describes DmDialog, a speech-to-speech dialogue translation system. DmDialog is one of the first experimental systems to perform speech-to-speech translation and the first to demonstrate the possibility of simultaneous interpretation. The hybrid parallel system integrates parallel marker-passing and connectionist networks. Other characteristics of the system include a simultaneous interpretation capability, mixed-initiative discourse understanding, cost-based ambiguity resolution and an integration of case-based and constraint-based processing. DmDialog is implemented and has been publicly demonstrated since March 1989. The current implementation translates Japanese into English and operates on the ATR's conference registration domain. Massively parallel implementations have been carried out on various machines and attained high performance.Part of the work reported here (the massively parallel implementation) is supported by the National Science Foundation under grant MIP-90/09109. I want to thank Hideto Tomabechi, Teruko Mitamura, Lori Levin, Masaru Tomita, Jaime Carbonell, Alex Waibel, James McClelland and Hitoshi Iida for fruitful discussions and continuing support; also, Testuya Higuchi for IXM2 implementation and Dan Moldovan and members of the SNAP project for SNAP implementations. Several anonymous referees for this journal offered helpful suggestions. I would like to express special thanks to Mitsuko Saito who trained me as a simultaneous interpreter. Without my intuitions as a simultaneous interpreter, this work would not have started.     

基于对抗学习的蒙汉神经机器翻译

在机器翻译模型的构建和训练阶段,为了缓解因端到端机器翻译框架在训练时采用最大似然估计原理导致的翻译模型的质量不高的问题,本文使用对抗学习策略训练生成对抗网络,通过鉴别器协助生成器的方式来提高生成器的翻译质量,通过实验选择出了更适合生成器的机器翻译框架Transformer,更适合鉴别器的卷积神经网络,并且验证了对抗式训练对提高译文的自然度、流利度以及准确性都具有一定的作用.在模型的优化阶段,为了缓解因蒙汉平行数据集匮乏导致的蒙汉机器翻译质量仍然不理想的问题,本文将Dual-GAN (dual-generative adversarial networks,对偶生成对抗网络)算法引入了蒙汉机器翻译中,通过有效的利用大量蒙汉单语数据使用对偶学习策略的方式来进一步提高基于对抗学习的蒙汉机器翻译模型的质量.     

题录信息的机器翻译方法

该文针对题录信息中的人名、地址、机构名和公司名的不同特征,分别设计了不同的翻译方法,并依靠词典和翻译规则,实现了大部分内容的翻译。对于人名翻译,该文设计了拼音转换、假名转换和同音转换的翻译方法;对于地址、机构名和公司名的翻译,该文提出了先切分、再翻译、最后调序的翻译流程。实验表明,利用该文的方法翻译人名、地址、机构名及公司名,能够取得不错的翻译效果。     

XML集成中关系数据的动态转换方法

在关系数据库向XML集成的过程中,静态转换方法对关系数据进行一次转换,无法满足实际应用中某些特定关系数据的转换需求。为实现动态转换,在静态转换系统通用映射模型的基础上提出动态转换系统,给出相应的动态转换算法。通过实例验证该算法的有效性和通用性。     

Metamodels for Computer-based Engineering Design: Survey and recommendations

The use of statistical techniques to build approximations of expensive computer analysis codes pervades much of today’s engineering design. These statistical approximations, or metamodels, are used to replace the actual expensive computer analyses, facilitating multidisciplinary, multiobjective optimization and concept exploration. In this paper, we review several of these techniques, including design of experiments, response surface methodology, Taguchi methods, neural networks, inductive learning and kriging. We survey their existing application in engineering design, and then address the dangers of applying traditional statistical techniques to approximate deterministic computer analysis codes. We conclude with recommendations for the appropriate use of statistical approximation techniques in given situations, and how common pitfalls can be avoided.     

Nine Issues in Speech Translation

This paper sketches research in nine areas related to spoken language translation: interactive disambiguation (two demonstrations of highly interactive, broad-coverage speech translation are reported); system architecture; data structures; the interface between speech recognition and analysis; the use of natural pauses for segmenting utterances; example-based machine translation; dialogue acts; the tracking of lexical co-occurrences; and the resolution of translation mismatches.     

SIP IP多媒体通信系统的地址翻译

对SIPIP多媒体通信系统进行了简单的介绍。详细分析了DNS系统中的SRV和NAPTR两种资源记录及其使用方法。重点描述了SIPIP多媒体通信系统中的地址翻译过程,包括客户端的地址翻译和服务器端的地址翻译。     

Some key issues in designing MT systems

Development of a machine translation system (Mts) requires many tradeoffs in terms of the variety of available formalisms and control mechanisms. The tradeoffs involve issues in the generative power of grammar, formal linguistic power and efficiency of the parser, manipulation flexibility for knowledge bases, knowledge acquisition, degree of expressiveness and uniformity of the system, integration of the knowledge sources, and so forth. In this paper we discuss some basic decisions which must be made in constructing a large system. Our experience with an operational English-Chinese Mts, ArchTran, is presented to illustrate decision making related to procedural tradeoffs.We would like to thank the Behavior Tech Computer Corp. for its full financial support for research described in this report, and members of the MT research team at the corporation's R&D center for technical support. Also, several referees for this journal offered helpful suggestions.     

基于网络搜索的英汉人名翻译

该文利用搜索引擎从网络中挖掘英语人名的中文翻译。该方法综合利用翻译辅助词、英中人名共现规则、音译相似度和翻译概率。首先,利用搜索引擎从互联网上搜索英文人名的中文翻译候选。把汉语人名标注结果、翻译辅助词、英中人名共现规则和英文人名的发音音节长度结合起来提取翻译候选词。翻译辅助词有助于搜索与英文人名更相关的信息,英中人名共现规则和发音音节长度进一步缩小英文人名的翻译范围,使得英文人名的翻译搜索符合人名共现规律和发音规律。然后,根据音译相似度和翻译概率对候选词进行排序。人名翻译的绝大部分是根据发音翻译过来的,音译相似度是帮助判断两个词在发音上的相似性。翻译概率从统计上判断两个词互为翻译的可能性。实验结果表明,翻译辅助词、规则、音译相似度和翻译概率都有助于提高人名翻译的正确率。     

一种基于统计和模板的双层翻译研究

为了提高翻译系统的翻译准确率,在短语基础上结合模板的方法自动抽取模板结构;解码时,首先进行模板匹配,套用模板结构进行翻译,然后再按照Beam Search搜索算法进行后续翻译。因此,该方法可以有效地解决单一的统计翻译中语序错误。以汉蒙翻译为例,实验结果显示此方法可以有效地提高翻译效果,翻译效率比基于短语的统计翻译方法提高10%。