Microsoft Translator Team Blog

News and Views from the Microsoft Translator Team in Microsoft Research.

Visit our Forums

For questions please use Microsoft Translator technical forums. We monitor them daily!

  • Microsoft Translator Team Blog

    Congratulations to the MSR-MT researchers!

    • 3 Comments

    The National Institute of Standards and Technology has published its 2008 Machine Translation Evaluation Results.  From their website:

    The NIST 2008 Machine Translation Evaluation (MT-08) is part of an ongoing series of evaluations of human language translation technology. NIST conducts these evaluations in order to support machine translation (MT) research and help advance the state-of-the-art in machine translation technology. These evaluations provide an important contribution to the direction of research efforts and the calibration of technical capabilities. The evaluation was administered as outlined in the official MT-08 evaluation plan.

    Check out our research team's results in Chinese to English - very impressive.  Congratulations, guys!  Very well-deserved. 

    The MSR-MT team is looking to build on this exciting progress and bring this technology to you. Stay tuned!  

    Lane

    (*edit to correct link - thanks for pointing it out!)

  • Microsoft Translator Team Blog

    New Blogger on the block – Please welcome Lane!

    • 3 Comments

    I wanted to post a quick note welcoming Lane, who will be bringing you updates (more regularly than before) about what’s new and exciting in our team. Welcome Lane!

    On that note, Lane and Andrea were at the MSR Silicon Valley Road Show event showing off our products and technology. Robert Scoble from Fastcompany has a video:

     

    While still on the topic of videos – some of you might have missed news coverage from CNET News a little while back. We are covered in the last one and half minutes of this video.

     

    -Vikram

    Vikram works on Incubation Strategy for Microsoft Research Machine Translation team
  • Microsoft Translator Team Blog

    Machine Translation group at Microsoft Research

    • 0 Comments

    Microsoft Research’s Machine Translation (MSR-MT) group has been among the leading research organizations in the machine translation space for over 8 years, and some of the foundational work in natural language processing at MSR began over 16 years ago. The team’s approach to machine translation integrates linguistic features with state-of-the-art statistical machine translation algorithms. The team’s focus has always been on automatically acquiring translation knowledge from bilingual corpora, i.e., parallel data consisting of original source language sentences and their corresponding translations by human translators. About 3 years ago, the team’s focus shifted from a purely rule-based approach to this task toward a hybrid approach that includes extensive statistical processing, allowing for greater scalability across domains and into new languages.

    Microsoft’s Machine Translation technology was first developed for in-house localization purposes, to allow our Customer Support organization to publish technical support documents with a frequency and language breadth that would have been prohibitively expensive using human translators.  With all of Microsoft’s previously human-translated documents and localized software at its disposal, the MT team was able automatically to train its statistical MT engine to achieve quite good quality in the technical domain.  This technology was extended to support the Windows localization team, the Developer Division, MSDN, and several other groups within Microsoft. It has also allowed Microsoft to reach many more customers than would have ever been possible using human translation alone.

    After focusing on Microsoft’s own translation needs, the team began to build a scalable web service that would allow it to provide translation services to the general public, as a standalone tool on the web, and as a feature within other products.  Given that the Microsoft MT engine has been trained most heavily on technical data, it has not yet been tuned for translating text in other subject domains.  However, we hope to continue improving the quality and breadth of the engine. We look forward to sharing our developments with you over the coming months on this blog.

    - Heather

    Heather Thorne is the Director of Business Strategy at the Microsoft Research Machine Translation group
  • Microsoft Translator Team Blog

    Multilingual App Toolkit 3.1 Launched – Localize Your App, Extend Your Reach

    • 0 Comments

    Today, the Multilingual App Toolkit (MAT) team announced the release of MAT 3.1. The Multilingual App Toolkit integrates with Microsoft Visual Studio to provide Windows Store apps and Windows Phone apps with translation support powered, translation file management, and editor tools powered by Microsoft Translator.

    This release provides several key fixes as well as new and improved features including:

    • Visual Studio Online builds
    • Expanded import, export and recycling
    • Improved translation and suggestion results
    • As well as a number of key fixes

    The MAT team really focused on key features as well as addressing both reported and non-reported issues in this release. A huge congratulations to the MAT team and their great work delivering this latest release!

    Please note that due to updates to the setup process, you will need to perform a one-time uninstall of MAT v3.0 or earlier before installing v3.1.

    Get Started Using MAT Today

  • Microsoft Translator Team Blog

    Language Can Make the Difference Between Good and Great Customer Service

    • 0 Comments

    The following is a guest post by the Microsoft Translator Partner, Lionbridge Technologies, who developed GeoFluent as solution to address the challenge of real-time translation of user generated content leveraging the Microsoft Translator automatic translation service and customization capabilities of the Translator Hub.

    ----------------------------------------

    It’s the middle of the night in China, and you are still at the office, working on a complex installation that needs to be up and running by the time the boss comes in the next day. There is no room for error, and the pressure is on.

    All of the sudden, a server malfunctions, and you realize your big project has stalled, and progress has come to a halt. What do you do?

    Your first lifeline is the vendor’s “Contact Us” button on their website. A chat window opens up immediately with customer support on the other end. An agent works through the issue with you online, and suddenly the project is back underway. You breathe a deep sigh of relief.

    What you don’t know is that the customer support agent only speaks English and was seamlessly interacting with you, even though your native language is Chinese. The communication was so seamless you weren’t even aware that a language gap existed. Somehow, it just worked.   What you did know, however, is that, had immediate assistance from customer support not been available, you would likely be looking for a new vendor for future business.

    The lesson learned is simple.  Today’s business demands that companies of all types need to be able to respond and support their global customers 24 hours a day, 7 days a week, 365 days a year.  That means that customer service personnel must be able to provide that support to customers regardless of their native language and time zone. 

    The challenge with that type of customer service and support is that neither native speakers or human translation are viable solution, both from a cost, response-time and scalability standpoint. Can machine translation solve this problem effectively?

    As experts in translation and localization services, we at Lionbridge understood that while automatic translation provided the immediacy necessary to keep up with the translation demands of user-generated content, business clients demanded more.  Translations had to at least be actionable and understandable even if they weren’t perfect.

    The list of challenges that needed to be addressed to make automatic translation valid for these customer support scenarios were numerous. Some of the biggest hurdles included:  

    1. Typos and Slang - With user-generated content, people tend to take shortcuts and write in a less formal and more conversational manner resulting in typos, slang, shorthand and industry-centric language. While this is common, it is problematic for translation systems to identify and fix appropriately.  
    2. Chat Client - Each customer had a customized version of the chat client that implies any translation solution needed to integrate into their existing workflow and products.  
    3. Global Search– Monolingual search in forum and community sites is much different than multilingual search from both a search criteria and results perspective.  Finding an approach that is intuitive for the end user had to be a primary focus.

    For this reason, we developed GeoFluent, a real-time automated translation solution designed to meet the needs of business clients in a scalable and cost effective manner. Building on Microsoft Translator as the  underlying automatic translation service, GeoFluent provides an additional configuration layer that enables real-time conversations and content sharing with users who speak, read and write different languages. It is easily integrated into customer’s existing
    chat and community applications. 

    We took these above mentioned challenges into account when developing GeoFluent. This SaaS-based application delivers the features that businesses require, including:

    • Masking personally identifiable information during the translation process.
    • Protecting company, brand and product names so that the right corporate image is delivered in any language.
    • Building company glossaries and systems to provide preferred translations to accommodate corporate, industry and locale-specific terminology through use of the Translator Hub.
    • Adding layers of linguistic processing to remove slang, typos, grammatical and punctuation errors from source material and provide support for regional variants of languages.

    A successful machine translation experience must be seamless, actionable, and timely. GeoFluent makes this makes this possible for the demanding needs of customer service organizations to support a global customer base.

    To learn more about how GeoFluent solutions are leveraging Microsoft Translator, go to:

    GeoFluent Video

    GeoFluent Partner Page

     

    By Greg Belkin

    Director, Product Marketing

    Lionbridge Technologies

    www.lionbridge.com

Page 16 of 21 (101 items) «1415161718»