This is the year of machine learning and big data. Whether it is predicting political results, supercharging your Excel spreadsheets, helping map queries to intent in Search, or even customizing a translation engine to best fit your content – these research areas are playing a starring role in transforming technology and productivity.
A couple of weeks back, at the 14th annual Computing in the 21st Century Conference, attendees saw a glimpse of where else these technologies are taking us – and loved it. Rick Rashid, who heads up Microsoft Research worldwide, went up on stage and in the span of eight sentences, got the 2000+ strong crowd up on their feet and cheering. It was a moment where technology was indistinguishable from magic – and one that would spur science fiction writers to start thinking of bigger challenges for researchers to tackle :)
Watch the video to see for yourself:
A combination of powerful technologies were employed to make this amazing demonstration possible: Deep Neural Network based processing combined with high performance computing allowed a significant jump in accuracy of speech recognition. The Microsoft Translator technology that you use each day was customized to best fit Rick’s speech content. New speech synthesis technology that allows personalization of acoustic characteristics was able to create “Rick’s voice” in a language he does not speak. You can read Rick’s blog post here.
Some of these technologies are already available today, especially the industry-leading translation (Microsoft Translator) with customization capabilities (Translator Hub). If you are a Windows Phone user, you have been enjoying the most innovative translation app on any phone for over a year now, which includes an early speech translation experience that has been tuned for travel situations. The audio output that you hear on Bing Translator website uses some of the newer speech synthesis engines coming out of our Speech research. Deep-Neural-Net research is also behind our audio/video indexing service – MAVIS, which is available commercially.
The excitement that has been rippling across the web in response to this demonstration is an indicator of how much everyone wants to experience this ‘magic’. There is much work to do, but you will see the benefits of this amazing research in our products in our future releases.
Vikram Dendi Director Microsoft/Bing Translator & Microsoft Research
Delivering free, easy-to-use tools to enable you and your community to collaboratively customize translations based on your content and scenarios.
As machine translation researchers, we are well aware of the challenges in applying brute force computing power to solve translation problems. We know that no matter how much processing power you throw at translation, it is still a stretch to get an error-free, contextually accurate translation every time. As a partner-focused translation services team, we have been on the forefront of delivering better ways to tailor translations to fit the specific content being translated. Over two years ago, we took a step in the direction of helping users customize translations being delivered through our Microsoft Translator Collaborative Translation Framework. As an integral part of the Microsoft Translator API, these technologies allowed users to edit and override the machine generated translations after they were delivered, and made them available for reuse via the API.
Today at Microsoft’s Worldwide Partner Conference, we are announcing the commercial availability of the Microsoft Translator Hub, an innovative tool that gives partners and communities unprecedented control over how the translation engine translates their content -- before the translations are delivered. Using the Hub, users can improve and optimize the translation quality for a specific area of terminology and style.
The Translator Hub is a free extension of the popular Microsoft Translator service, and enables businesses to combine existing translated documents with the power of Microsoft Translator’s big data backend to easily build a custom translation system, whose quality is controlled by the business. Custom systems built and deployed are seamlessly accessible via the standard Microsoft Translator API, and can be built into any scenario or workflow.
While the technology behind the translation and customization services is very powerful, our goal was to deliver the Hub as a simple to use private web portal that makes it easy for users to get started quickly. We achieved this by enabling users to build custom machine translation systems in four simple steps.
The users of the Hub can upload parallel (same document in two languages) and monolingual (single language) documents in a variety of formats, and build custom translation models in a private workspace using Microsoft Translator’s machine learning based training systems. The Hub provides methods and a simple user interface for collaborating and improving the translation system with reviewers, before deploying to the Microsoft Translator runtime infrastructure. The owner of the customized system can keep the system private, share it with other individuals, companies, or make it available publicly.
In addition the same collaborative translation functionality is integrated into the Microsoft Translator API enabling continuous improvement of the customized translation system through ongoing community engagement and feedback.
Learn more about this great tool on the Microsoft Translator web site, where you can also see how some of our early partners, like Lionbridge and PLYmedia, have leveraged the Translator Hub to power innovative business solutions and scenarios. You can also request an invite to the Hub directly from the Translator Hub portal.
We are confident that this technology will change the conversation about the quality of machine translation. Whether you are looking to stretch your localization budget, communicate with your global customers, or better understand your increasingly multilingual business data, Microsoft Translator Hub and the Translator API are worth considering as part of your workflow. By bringing together your pre-existing translated data with Microsoft’s big data translation models, the Hub opens up new cross-language possibilities for your business.
We look forward to working with you. If you are attending WPC 2012, do attend the Microsoft Translator session (2 PM, Wednesday July 11) or visit our innovation theatre presentations in the Solutions Innovation Center to learn more (search for “microsoft translator”).
- Vikram Dendi Director, Product Management Microsoft/Bing Translator
In partnership with Microsoft Research Connections, we also had the privilege of showcasing another aspect of the Microsoft Translator Hub in helping preserve and revitalize languages online in February 2012. Members of the Hmong community were among the first users of the Translator Hub and were able to build a machine translation system for the Hmong Daw language from scratch. The community chose to make this language available broadly via the public translation API and Bing Translator on International Mother Language Day, helping the worldwide Hmong community benefit from the great work of these passionate volunteers. Many other communities from around the world are now using the Translator Hub to build translation systems for their languages. You can watch some of these inspiring stories here and learn more about the research behind the Microsoft Translator Hub on the Inside Microsoft Research blog.
We are pleased to welcome Yahoo! Babel Fish users to the Bing Translator family. We have been working closely with our friends at Yahoo! to make this an easy transition, and Bing Translator is a natural upgrade to the experience with Yahoo! Babel Fish. We support all the languages you used with Babel Fish, and provide a superset of all the features.
You will notice a welcome banner indicating your transition to the Bing Translator site from a Yahoo site when you arrive at Bing Translator. I am pleased to introduce to you some of the great features that our translation service offers starting with a brief introduction to the technology behind the service.
Our translation technology is built on over a decade of world-class work done at Microsoft Research, and is widely used by a variety of applications and websites. You may have already encountered our technology when you used the translation features within products like Microsoft Office, Bing, Windows Phone, Facebook and Kindle. We serve billions of translations every day across 38 different languages, and we continually add to the list of languages and features. We also have a fast growing developer community.
Text translator: You can translate text snippets between the supported languages on the home page. If you aren’t sure about the language of the source text, you can always leave it as “Auto-Detect” and we will detect it for you. You can also listen to the translation for a sub set of languages by clicking the speaker icon. You can search the web with translated text snippet, or email it. We welcome your feedback on the translation quality.
Webpage translator: Simply type a webpage URL into the text entry box on the home page to translate a webpage. You can use the bi-lingual viewer functionality and switch between various views, including the popular side-by-side view. As you navigate through the site, the webpage translator continues translating.
Website widget: We delivered the first no-code, in-place translation widget several years ago and have improved it further since then. The Microsoft Translator widget is a simple to use web control that can make your site available in multiple languages without any coding, while keeping your users on your site. The unique collaborative translation functionality allows you to customize the translations delivered to your site by partnering with trusted users and allow your community to provide suggestions and corrections, to ensure the translations are always best suited to your content and always up-to-date.
You may also find it interesting to explore the language labs, where you can test new prototypes and demos from the minds of our team’s researchers and engineers. Our blog is here, you can follow us on Twitter or Facebook and if you have any questions or need any technical support you can use the forums.
You can also directly reach out to us at email@example.com if you have any other questions.
Once again, welcome! We are very pleased to have you.
Vikram Dendi, Director, Bing/Microsoft Translator (@vikman)
Updated (5/30): Fixed some links
Something amazing happened as GPS navigation devices started making their way into the hands and cars of people around the world – it created a new band of adventurer. Road maps were discarded, atlases gathered dust and schedules filled with more destinations. The GPS gave it’s user that bit of extra confidence to boldly go where they might have otherwise had to spend time to prepare to go.
With the new Translator app for Windows Phone, you will be well equipped for an adventure of your own – especially in going to places where you don’t read or speak the local language. With the app on hand, translating printed language on street signs, posters, transit schedules, restaurant menus, etc., is a snap. Well, easier than a snap – all you do is point and observe! The brilliant video mode translation equips you with the magical experience of just looking through the camera to see everything continuously translated, nicely overlaid over the original language.Think automatic subtitles for everyday life. Able to work entirely offline after downloading a small language pack, this feature relies on a highly optimized and compressed translation system, one built using the same technology that powers Bing’s other translation features.
Here is a behind the scenes look at the app:
You can learn more about the functionality of the app on the Bing blog. If you are an existing user of a previous version, you should shortly see a notification to update.Translator for Windows Phone is now available completely free, on the Windows Phone Marketplace. We would love to hear your feedback and feature suggestions on our user forums.
In building this app, engineers and researchers on our team solved tremendous technical challenges, and dealt with the nuances of voice, cameras, data availability and language complexity. Yet, our focus was on delivering a scenario that was truly useful to our users. If we are able to provide you that little bit of extra confidence that makes the difference between going somewhere and not – then we would have succeeded.
- Vikram Dendi, Director of Product Management, Microsoft/Bing Translator
· In celebration of International Mother Language Day, we are pleased to announce the addition of the Hmong language to our list of supported languages, made possible by a close partnership with the Hmong community. Anyone can now try out the new language on the Bing Translator site, or call it via the Microsoft Translator web service (Hmong Daw, language code mww). Hmong Daw is the dialect of Hmong the system supports, also known as White Hmong.
Instrumental to this effort were members of the Hmong community, who were able to leverage new tools from Microsoft Translator to help preserve and revitalize their language online. These new tools, currently in beta, enable automatic translation support for additional languages, or building higher quality systems for specific terminology and style in the established languages.
The addition of the Hmong language is an example of the first scenario: Members of the community utilized existing translated material and new features of Microsoft Translator to train a new translation engine. This leveraged Microsoft Translator’s learning abilities, which can learn how to translate from a set of parallel documents (same document in two languages), dictionaries and texts in the language to translate to (Hmong in this case). In addition to teaching the engine a new language, they also involved members of the community, partners and collaborators to create and review improved versions of the automated translation system, and collect qualitative feedback about each “trained” system. Deploying a system that reaches a certain level of quality allows seamless use with the standard Microsoft Translator APIs, and many scenarios powered by the API, like the web translation widget. Feedback that is generated through these scenarios can be utilized again in the training process – creating a virtuous loop for improving the translation quality. Stay tuned for more details about these new tools/features.
Once again, on International Mother Language day, we congratulate the Hmong community on their accomplishment. We are looking forward to working with many more partners and language communities in the near future.