DOCSLIB.ORG

The MT Developer/Provider and the Global Corporation

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
The MT Developer/Provider and the Global Corporation The MT developer/provider and the global corporation Terence Lewis1 & Rudolf M.Meier2 1 Hook & Hatton Ltd, 2 Siemens Nederland N.V. [email protected], [email protected] Abstract. This paper describes the collaboration between MT developer/service provider, Terence Lewis (Hook & Hatton Ltd) and the Translation Office of Siemens Nederland N.V. (Siemens Netherlands). It involves the use of the developer’s Dutch-English MT software to translate technical documentation for divisions of the Siemens Group and for external customers. The use of this language technology has resulted in significant cost savings for Siemens Netherlands. The authors note the evolution of an ‘MT culture’ among their regu- lar users. 1. Introduction 2. The Developer’s perspective In 2002 Siemens Netherlands and Hook & Hat- Under the current arrangement, the Developer ton Ltd signed an exclusive agreement on the (working from home in the United Kingdom) marketing of Dutch-English MT services in the currently acts as the sole operator of an Auto- Netherlands. This agreement set the seal on an matic Translation Environment (Trasy), process- informal partnership that had evolved over the ing documents forwarded by e-mail by the Trans- preceding six-seven years between Siemens and lation Office of Siemens Netherlands in The the Developer (Terence Lewis, trading through Hague. Ideally, these documents are in MS Word the private company Hook & Hatton Ltd). Un- or rtf format, but the Translation Office also re- der the contract, Siemens Netherlands under- ceives pdf files or OCR output files for automatic takes to market the MT Services provided by translation. After conversion to a machine-pro- the Developer to divisions within the Siemens cessable format, these files need to be cleaned Group and to prospective external customers. up, otherwise the MT output will be markedly This arrangement has proved to be a win- inferior to the agreed baseline. This clean-up is win situation for both Siemens Netherlands and done by either party by arrangement. the Developer. It has given Siemens Nether- The documents are processed and returned lands access to translation technology at a time on the basis of an agreed delivery date. Short when it has been involved in major cross-border documents are often returned within the hour. infrastructure projects with consortium partners They are not post-edited or revised line by line, that need rapid access to usable English transla- but the Developer does quickly scan each page tions of Dutch technical and legal documenta- to spot any gross errors or omissions. Sentences tion (technical requirements, relevant legisla- containing gross errors are usually sent back to tion and regulations etc.). Siemens is responsi- the MT system after further preparation work. ble for the complete administration of the MT Broadly speaking, users receive automatically customers, and this set-up has provided the De- generated documents. These are either edited by veloper with a stable stream of revenue ena- the end users themselves or used “as is” by en- bling him to devote more time to further devel- gineers familiar with the technical background opment of the software. This paper provides a to the documents. mostly practical account of this relationship, In addition to the run-of-the-mill jobs that first through the eyes of the Developer, and pass through a corporate translation office, the then from the perspective of the global corpora- automatic translation environment has been tion. used to handle huge volumes of documentation EAMT 2005 Conference Proceedings 176 The MT developer/provider and the global corporation from such major projects as the HSL-Zuid (the nal delivery tool and only sending to the MT high-speed rail link between Amsterdam and engine sentences that are not already present in Paris), the RandstadRail light-rail project in the the translation memory database. The MT soft- Hague Region, and a huge construction project ware produces a file that can imported directly for an external customer, Shell International. into the translation memory. This selection, ex- Trasy is built around a small LAN compris- traction and integration is performed semi- ing 4 PCs: automatically. Every job the Developer does is ultimately stored in the translation memory da- • a Linux gateway to the Internet tabase. Figure 1 shows an example of transla- • two Linux translation servers tion unit in an aligned file produced by the MT • a translator workstation/client running un- software. It should be noted that the MT soft- der Windows 98 ware can also be used on a stand-alone basis, This configuration has been set up to facilitate producing an html output file. the handling of a flow of jobs in an office envi- ronment. Of course, the number of translation <TrU> servers and workstation/clients can be increased <Quality>98 to meet any required number of translator/ope- <CrU>LEWIS! rators. All the dictionary building work is done <CrD>15042004,08:38:04 on the translator workstation and the up-to-date <Seg L=NL_NL>De hoofddraagconstructie is van lexical resources are then uploaded together with staal. the files to be translated to the Linux servers. <Seg L=EN-GB>The main supporting structure is of Typically, the Developer/Provider will start the steel. terminology work for the next job while the cur- </TrU> rent job is being processed. This client/server Figure 1: Translation unit produced by MT software arrangement has been found to the most practi- cal one, but it is possible (although somewhat The following software tools are used in this inconvenient) to put the whole set-up on a pow- translation environment: erful notebook running either Linux (or Solaris) or Windows XP (or both) that functions as a • Resource Building Tools mobile translation workstation. • Dutch-English machine translation software The use of two translation servers by a sin- • Trados Translator’s Workbench gle operator makes it easy to jump the queue to The Resource Building Tools, written by the handle shorter jobs when a long job is already Developer, include conventional single-word dic- running. It also enables the Developer to run the tionary-building tools plus tools for building re- same job simultaneously on two successive builds sources consisting of multiword expressions up of the software, thus providing a simple mecha- to idiomatic translations of complete sentences. nism for observing the success or failure of in- The utility called “ResourceBuilder” can pro- tended software improvements. This is vital since duce useful sub-sentence segments from trans- the Dutch-English MT application, written en- lation memory databases, aligned source/target tirely in Java, is seen as a dynamic process – files and even different language versions of most large jobs will generate at least one im- web pages. It is the key factor in securing user provement in the software. And the process of acceptance of the automatically generated improvement has been customer-driven: mis- translations since many of the constituent parts takes that were tolerated six years ago are now of sentences are taken from “human transla- pointed out with alarm. tions”. The single-word dictionary and the Re- sources Repository are basically simple re- 3. Software tools source files (enhanced text files) which are only All jobs are processed using the already de- compiled at run-time. The Resources Repository scribed approach combining Machine Transla- contains all lexical items that are not single- tion and Translation Memory. This involves us- word entries. ing a translation memory application as the fi- Examples of entries in this repository are: EAMT 2005 Conference Proceedings 177 Lewis and Meier “Tot het werk behoort”, “<MF>The work and, with sufficient data, can produce complete includes</MF>„ translations independently within limited do- (literal translation =„to the work belongs”) mains. Figure 3 shows the translation of a com- “van den Berg”, “<N1 semType= plete sentence produced by mining the Re- „proper_noun”>van den Berg</N1>„ sources Repository with one word being looked (Dutch surname, not “of the Mountain”!) up in the single-word dictionary. After this module has run, the sentence is “in dit document wordt beschreven”, “<MF>this then passed to a Rule-Based Machine Transla- document describes</MF>„ tion module which applies various known MT (literal translation: “in this document is described”) techniques to the sentence. These include the “Centrale Velsen”, “<N1>Velsen Power parsing of “clusters” or chunks of text, recogni- Plant</N1>„ tion of known patterns and techniques for “certificerende instantie van de opdrachtgever, achieving semantic disambiguation based on “<N1>client’s certification agency</N1>„ the identification of the subject-matter of the (literal translation: “certifying instance of the document. At this stage, for instance, the string client”) “niet zal kunnen worden gekozen” would al- “Dit systeem is gelaagd opgebouwd”, “<MF>this ready be identified and tagged as a verbal phrase system has been built up in layers</MF>„ and the provisional translation “will not be able (literal translation: “This system is layered built to be chosen” would be allocated for it (the up”) word-for-word translation being “not will can be chosen”). The term “provisional” is used Figure 2: Examples of repository entries since other modules might “decide” to change The Dutch-English translation software is a mo- this translation at a later stage. These various dular application incorporating a hybrid approach techniques are applied “sentence down”, i.e. the to machine translation. Since the Developer be- last series of rules (contained in an Issu- lieves that this approach is critical in achieving eResolver module) deals with single words. In an acceptable quality level for Siemens Nether- the final part of the ‘run-time sequence’, the lands users, a few paragraphs will be devoted to sentence is passed to the EnglishStyleEditor the internal flow of the application.
Load more

Recommended publications
  • Neuro Informatics 2020
    Neuro Informatics 2019 September 1-2 Warsaw, Poland PROGRAM BOOK What is INCF? About INCF INCF is an international organization launched in 2005, following a proposal from the Global Science Forum of the OECD to establish international coordination and collaborative informatics infrastructure for neuroscience. INCF is hosted by Karolinska Institutet and the Royal Institute of Technology in Stockholm, Sweden. INCF currently has Governing and Associate Nodes spanning 4 continents, with an extended network comprising organizations, individual researchers, industry, and publishers. INCF promotes the implementation of neuroinformatics and aims to advance data reuse and reproducibility in global brain research by: • developing and endorsing community standards and best practices • leading the development and provision of training and educational resources in neuroinformatics • promoting open science and the sharing of data and other resources • partnering with international stakeholders to promote neuroinformatics at global, national and local levels • engaging scientific, clinical, technical, industry, and funding partners in colla- borative, community-driven projects INCF supports the FAIR (Findable Accessible Interoperable Reusable) principles, and strives to implement them across all deliverables and activities. Learn more: incf.org neuroinformatics2019.org 2 Welcome Welcome to the 12th INCF Congress in Warsaw! It makes me very happy that a decade after the 2nd INCF Congress in Plzen, Czech Republic took place, for the second time in Central Europe, the meeting comes to Warsaw. The global neuroinformatics scenery has changed dramatically over these years. With the European Human Brain Project, the US BRAIN Initiative, Japanese Brain/ MINDS initiative, and many others world-wide, neuroinformatics is alive more than ever, responding to the demands set forth by the modern brain studies.
    [Show full text]
  • Three Directions for the Design of Human-Centered Machine Translation
    Three Directions for the Design of Human-Centered Machine Translation Samantha Robertson Wesley Hanwen Deng University of California, Berkeley University of California, Berkeley samantha [email protected] [email protected] Timnit Gebru Margaret Mitchell Daniel J. Liebling Black in AI [email protected] Google [email protected] [email protected] Michal Lahav Katherine Heller Mark D´ıaz Google Google Google [email protected] [email protected] [email protected] Samy Bengio Niloufar Salehi Google University of California, Berkeley [email protected] [email protected] Abstract As people all over the world adopt machine translation (MT) to communicate across lan- guages, there is increased need for affordances that aid users in understanding when to rely on automated translations. Identifying the in- formation and interactions that will most help users meet their translation needs is an open area of research at the intersection of Human- Computer Interaction (HCI) and Natural Lan- guage Processing (NLP). This paper advances Figure 1: TranslatorBot mediates interlingual dialog us- work in this area by drawing on a survey of ing machine translation. The system provides extra sup- users’ strategies in assessing translations. We port for users, for example, by suggesting simpler input identify three directions for the design of trans- text. lation systems that support more reliable and effective use of machine translation: helping users craft good inputs, helping users under- what kinds of information or interactions would stand translations, and expanding interactiv- best support user needs (Doshi-Velez and Kim, ity and adaptivity. We describe how these 2017; Miller, 2019). For instance, users may be can be introduced in current MT systems and more invested in knowing when they can rely on an highlight open questions for HCI and NLP re- AI system and when it may be making a mistake, search.
    [Show full text]
  • TAUS Speech-To-Speech Translation Technology Report March 2017
    TAUS Speech-to-Speech Translation Technology Report March 2017 1 Authors: Mark Seligman, Alex Waibel, and Andrew Joscelyne Contributor: Anne Stoker COPYRIGHT © TAUS 2017 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system of any nature, or transmit- ted or made available in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of TAUS. TAUS will pursue copyright infringements. In spite of careful preparation and editing, this publication may contain errors and imperfections. Au- thors, editors, and TAUS do not accept responsibility for the consequences that may result thereof. Design: Anne-Maj van der Meer Published by TAUS BV, De Rijp, The Netherlands For further information, please email [email protected] 2 Table of Contents Introduction 4 A Note on Terminology 5 Past 6 Orientation: Speech Translation Issues 6 Present 17 Interviews 19 Future 42 Development of Current Trends 43 New Technology: The Neural Paradigm 48 References 51 Appendix: Survey Results 55 About the Authors 56 3 Introduction The dream of automatic speech-to-speech devices enabling on-the-spot exchanges using translation (S2ST), like that of automated trans- voice or text via smartphones – has helped lation in general, goes back to the origins of promote the vision of natural communication computing in the 1950s. Portable speech trans- on a globally connected planet: the ability to lation devices have been variously imagined speak to someone (or to a robot/chatbot) in as Star Trek’s “universal translator” to negoti- your language and be immediately understood ate extraterrestrial tongues, Douglas Adams’ in a foreign language.
    [Show full text]
  • Translation Technology Landscape Report
    Translation Technology Landscape Report April 2013 Funded by LT-Innovate Copyright © TAUS 2013 1 Translation Technology Landscape Report COPYRIGHT © TAUS 2013 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system of any nature, or transmitted or made available in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of TAUS. TAUS will pursue copyright infringements. In spite of careful preparation and editing, this publication may contain errors and imperfections. Authors, editors, and TAUS do not accept responsibility for the consequences that may result thereof. Published by TAUS BV, De Rijp, The Netherlands. For further information, please email [email protected] 2 Copyright © TAUS 2013 Translation Technology Landscape Report Target Audience Any individual interested in the business of translation will gain from this report. The report will help beginners to understand the main uses for different types of translation technology, differentiate offerings and make informed decisions. For more experienced users and business decision-makers, the report shares insights on key trends, future prospects and areas of uncertainty. ,QYHVWRUVDQGSROLF\PDNHUVZLOOEHQH¿WIURPDQDO\VHVRIXQGHUO\LQJYDOXH propositions. Report Structure This report has been structured in discrete chapters and sections so that the reader can consume the information needed. The report does not need to be read from VWDUWWR¿QLVK7KHFRQYHQLHQFHDIIRUGHGIURPVXFKDVWUXFWXUHDOVRPHDQVWKHUHLV some repetition. Authors: Rahzeb Choudhury and Brian McConnell Reviewers - Jaap van der Meer and Rose Lockwood Our thanks to LT-Innovative for funding this report. Copyright © TAUS 2013 3 Translation Technology Landscape Report CONTENTS 1. Translation Technology Landscape in a Snapshot 8 2.
    [Show full text]
  • Download Your FREE Copy of the LT2013
    THE FORUM FOR EUROPE'S LANGUAGE TECHNOLOGY INDUSTRY Language Technologies LT2013: Status and Potential of the European Language Technology Markets March 2013 www.lt‐innovate.eu ‐ contact@lt‐innovate.eu www.lt‐innovate.eu ‐ contact@lt‐innovate.eu TABLE OF CONTENTS Table of Contents .....................................................................................................................................1 Executive Summary ..................................................................................................................................2 Key Extracts....................................................................................................................................3 1. A New ICT Ecosystem ...........................................................................................................................6 1.1 The deep transformation of the competitive landscape .........................................................6 1.2 A Mobile/Social/Global Ecosystem ..........................................................................................7 2. Markets for Language Technology ......................................................................................................11 2.1 The Market Context for LT ......................................................................................................11 2.2 The LT‐Innovate Market Model ...............................................................................................14 2.3 Trends & Growth by Segment.................................................................................................17
    [Show full text]
  • Problem Solving Activities in Post-Editing and Translation from Scratch a Multi-Method Study
    Problem solving activities in post-editing and translation from scratch A multi-method study Jean Nitzke language Translation and Multilingual Natural science press Language Processing 12 Translation and Multilingual Natural Language Processing Editors: Oliver Czulo (Universität Leipzig), Silvia Hansen-Schirra (Johannes Gutenberg-Universität Mainz), Reinhard Rapp (Johannes Gutenberg-Universität Mainz, Hochschule Magdeburg-Stendal) In this series: 1. Fantinuoli, Claudio & Federico Zanettin (eds.). New directions in corpus-based translation studies. 2. Hansen-Schirra, Silvia & Sambor Grucza (eds.). Eyetracking and Applied Linguistics. 3. Neumann, Stella, Oliver Čulo & Silvia Hansen-Schirra (eds.). Annotation, exploitation and evaluation of parallel corpora: TC3 I. 4. Czulo, Oliver & Silvia Hansen-Schirra (eds.). Crossroads between Contrastive Linguistics, Translation Studies and Machine Translation: TC3 II. 5. Rehm, Georg, Felix Sasaki, Daniel Stein & Andreas Witt (eds.). Language technologies for a multilingual Europe: TC3 III. 6. Menzel, Katrin, Ekaterina Lapshinova-Koltunski & Kerstin Anna Kunz (eds.). New perspectives on cohesion and coherence: Implications for translation. 7. Hansen-Schirra, Silvia, Oliver Czulo & Sascha Hofmann (eds). Empirical modelling of translation and interpreting. 8. Svoboda, Tomáš, Łucja Biel & Krzysztof Łoboda (eds.). Quality aspects in institutional translation. 9. Fox, Wendy. Can integrated titles improve the viewing experience? Investigating the impact of subtitling on the reception and enjoyment of film using eye tracking and questionnaire data. 10. Moran, Steven & Michael Cysouw. The Unicode cookbook for linguists: Managing writing systems using orthography profiles. 11. Fantinuoli, Claudio (ed.). Interpreting and technology. 12. Nitzke, Jean. Problem solving activities in post-editing and translation from scratch: A multi-method study. ISSN: 2364-8899 Problem solving activities in post-editing and translation from scratch A multi-method study Jean Nitzke language science press Nitzke, Jean.
    [Show full text]
  • A Retrospective and Prospective View Of
    A retrospective and prospective view of translation research from an empirical, experimental, and cognitive perspective: The International Journal for Translation & Interpreting Research the TREC network trans-int.org Amparo Hurtado Albir Universitat Autònoma de Barcelona [email protected] Fabio Alves Universidade Federal de Minas Gerais [email protected] Birgitta Englund Dimitrova Stockholms Universitet [email protected] Isabel Lacruz Kent State University [email protected] DOI: ti.106201.2015.a02 Abstract. The aim of this paper is to discuss some developments in empirical translation research with an experimental and cognitive perspective. The focus is on the activities and research of the network TREC (Translation, Research, Empiricism, Cognition). The network was formed in 2011, funded by the Spanish Ministry of Economy and Competitiveness and led by PACTE (Universitat Autònoma de Barcelona). It consists of translation scholars and research groups united by their common interest in empirical and experimental research, particularly in relation to the cognitive operations that underlie the task of translating. The paper first gives a short general overview of research on translation as a cognitive activity and outlines the objectives of the TREC network. The network members, representing universities from Argentina, Brazil, Denmark, Finland, Ireland, Norway, Spain, Sweden, Switzerland, the UK, and the USA, then present their most important contributions to cognitively oriented research (topics, methods, results). Finally, some conclusions are drawn and perspectives for future research are outlined. Keywords: translation, cognition, empirical-experimental research, process-oriented translation studies, networking 1. Research on translation as a cognitive activity Studies concerning translation as a cognitive activity date back to the late 1960s.
    [Show full text]
  • A Linguistics Study of New Trends in Translation As a Speech to Act
    International Journal of Science and Research (IJSR) ISSN (Online): 2319-7064 Index Copernicus Value (2013): 6.14 | Impact Factor (2013): 4.438 A Linguistics Study of New Trends in Translation as a Speech to Act Riyad Sarhan Jabbar Faculty of Education, University of Kufa, Najaf, Iraq, P. O. Box 21 Abstract: Machine translation is the only available alternative to human translation. Thus, evaluating machine translation software, is very important to improve the performance of these software. This study aims to highlight the difference between human translation and machine translation with reference to speech to text program . In this study , the approaches of translation are mentioned as direct approach , interlingua approach , and the transfer approach . The problem of this program is with the literary translation more than the other types of translation as scientific and idioms . The most accurate is the scientific one as this study proves . As a solution to overcome the problem of literary translation is that machine translation software should be modified and updated continuously , so that they can keep pace with the follow of literary terms . Also , the study shows the advantages and disadvantages of using machine translation. Moreover, the work illustrates the types of linguistic problems encountered in machine translation, and some solutions to overcome these problems. Keywords: Translation, Machine Translation, Translation software, speech, Text, Transfer approach, Interlingua approach 1. Introduction 1.2 The Aims of the Study 1.1 The Problems of the Study 1) The study will attempt to come out with some solutions to machine translation (MT) problem whenever possible The mechanization of translation it has been one of so as to come with highly quality automatic translation .
    [Show full text]
  • Automatic Text Extraction and Translation
    Image-Based Mobile Service: Automatic Text Extraction and Translation J´erˆome Berclaza, Nina Bhattib,StevenJ.Simskec and John C. Schettinob a CVLab, EPFL, Lausanne, Switzerland b HP Labs, Palo Alto, CA, USA c HP Labs, Fort Collins, CO, USA ABSTRACT We present a new mobile service for the translation of text from images taken by consumer-grade cell-phone cameras. Such capability represents a new paradigm for users where a simple image provides the basis for a service. The ubiquity and ease of use of cell-phone cameras enables acquisition and transmission of images anywhere and at any time a user wishes, delivering rapid and accurate translation over the phone’s MMS and SMS facilities. Target text is extracted com- pletely automatically, requiring no bounding box delineation or related user intervention. The service uses localization, binarization, text deskewing, and optical character recognition (OCR) in its analysis. Once the text is translated, an SMS message is sent to the user with the result. Further novelties include that no software installation is required on the handset, any service provider or camera phone can be used, and the entire service is implemented on the server side. Keywords: Scene text extraction, mobile phone camera, client / server, translation 1. INTRODUCTION Imagine a traveler dining in a restaurant in a country on the other side of the world and struggling with the menu. “Is it Beef Steak or Lizard Steak?” could be one of the more pressing questions as the traveler reads through the menu. What if this traveler could use her camera phone to get a translation simply by taking a picture of the menu, sending a MMS and promptly receiving an SMS with the answer? In this paper we describe such a system that uses any camera phone and any mobile service provider to accomplish this task quickly and easily.
    [Show full text]
  • Trkic G00gle: Why and How Users Game Translation Algorithms
    Trkic G00gle: Why and How Users Game Translation Algorithms SOOMIN KIM, Seoul National University, South Korea CHANGHOON OH, Boston College, USA WON IK CHO, Seoul National University, South Korea DONGHOON SHIN, Seoul National University, South Korea BONGWON SUH, Seoul National University, South Korea JOONHWAN LEE∗, Seoul National University, South Korea Individuals interact with algorithms in various ways. Users even game and circumvent algorithms so as to achieve favorable outcomes. This study aims to come to an understanding of how various stakeholders interact with each other in tricking algorithms, with a focus towards online review communities. We employed a mixed-method approach in order to explore how and why users write machine non-translatable reviews as well as how those encrypted messages are perceived by those receiving them. We found that users are able to find tactics to trick the algorithms in order to avoid censoring, to mitigate interpersonal burden, toprotect privacy, and to provide authentic information for enabling the formation of informative review communities. They apply several linguistic and social strategies in this regard. Furthermore, users perceive encrypted messages as both more trustworthy and authentic. Based on these findings, we discuss implications for online review community and content moderation algorithms. CCS Concepts: • Human-centered computing ! User studies. Additional Key Words and Phrases: Human-AI Interaction; algorithmic experience; gaming; translation algorithm; online review; recommendation algorithm; peer-to-peer platform ACM Reference Format: Soomin Kim, Changhoon Oh, Won Ik Cho, Donghoon Shin, Bongwon Suh, and Joonhwan Lee. 2021. Trkic G00gle: Why and How Users Game Translation Algorithms. Proc. ACM Hum.-Comput. Interact.
    [Show full text]
  • Unmet Needs and Opportunities for Mobile Translation AI Daniel J
    Unmet Needs and Opportunities for Mobile Translation AI Daniel J. Liebling, Abigail Evans Aaron Donsbach, Michal Lahav Northeastern University Jess Holbrook, Boris Smus, Google Seattle, WA Lindsey Boran Seattle, WA [email protected] Google dliebling,[email protected] Seattle, WA donsbach,jessh,smus,[email protected] ABSTRACT affordances including speech recognition and augmented real- Translation apps and devices are often presented in the con- ity [14]. text of providing assistance while traveling abroad. However, Access to translation, then, appears to be universally available. the spectrum of needs for cross-language communication is However, little public research exists that details people’s trans- much wider. To investigate these needs, we conducted three lation needs and examines how translation apps meet those studies with populations spanning socioeconomic status and needs. For what populations, in which scenarios, and what con- geographic regions: (1) United States-based travelers, (2) mi- tent is translated? How do translation apps succeed or fail at grant workers in India, and (3) immigrant populations in the these tasks? This paper sheds new light on these questions. We United States. We compare frequent travelers’ perception detail the language needs of, and use of translation technology and actual translation needs with those of the two migrant by, three different communities: English-speaking travelers communities. The latter two, with low language proficiency, based in the United States, Hindi-speaking intra-national mi- have the greatest translation needs to navigate their daily lives. grant workers who live in Tamil-speaking Chennai, India, and However, current mobile translation apps do not meet these immigrants with low English proficiency who live in Seattle, needs.
    [Show full text]
  • Development of Image Processing and Vision Systems with Industrial
    Development of Image Processing and Vision Systems with Industrial Applications Zhang Yi A THESIS SUBMITTED FOR THE DEGREE OF DOCTOR OF PHILOSOPHY DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING NATIONAL UNIVERSITY OF SINGAPORE 2009 Acknowledgments I would like to express my sincerest appreciation to all who had helped me during my study in National University of Singapore. First of all, I would like to thank my supervisors Associate Professor Tan Kok Kiong for his inspirational discussions, support and encouragement. His vision and passion for research enlighten my research work and spurred my creativity. I would like to give my gratitude to all my friends in Mechatronics and Automation Lab. I would especially like to thank Dr. Huang Sunan, Dr. Tang Kok Zuea, Mr. Tan Chee Siong, Dr. Zhao Shao, Dr. Teo Chek Sing, Dr. Andi Sudjana Putra, Mr. Chen Silu and Mr. Yuan Jian for their helpful discussions and advice. I would also wish to thank Ms Lay Geok from Medical Department, NUS for her assistance for my experiment. Finally, I would like to thank my family for their endless love and support. II CONTENTS Acknowledgments...................................................................................................II List of Figures ....................................................................................................... IV List of Tables ........................................................................................................ VI List of Abbreviations ...........................................................................................VII
    [Show full text]