Conversation Translator Auto-Translates Instant Messages Into 35 Languages

REDMOND, Wash. – June 20, 2011 – Need to ping a colleague overseas who doesn’t speak your language? Struggling to find the right word in English?

Enter Conversation Translator, a new add-on for Microsoft Lync that uses Bing Translator to automatically translate instant messages directly inside Lync. Developed by Microsoft Research (MSR) and the Microsoft Lync team, the service that launched on June 3 translates incoming and outgoing instant messages into 35 languages.

Conversation Translator is a new add-on that uses the Bing Translator to automatically translate instant messages directly inside Microsoft Lync.

Communicating without having to thinking about translation is a big step forward for communication, says Takako Aikawa, program manager on MSR’s Machine Translation Team. “The notion of being language-agnostic is essential for global communication, and that’s the beauty of this translator,” Aikawa says. “The goal is to let everyone communicate in their own language, and make every language – and every person – equal.”

Lin Li, senior test engineer for the Lync team and a researcher on language learning, sees Conversation Translator as a groundbreaking initiative at Microsoft. She believes that as social features become more prevalent in Microsoft products, translation will become ever more important.

“Global economic development brings language diversity and translation to the forefront, and as worldwide communication becomes more commonplace we’ll have more opportunities to offer solutions to our customers to help them address language barriers,” Li says. “So this translation feature – not only in Lync but across our products – will become more and more important.”

Language translation isn’t new to Microsoft products, Aikawa notes. It’s been integrated into Microsoft Office, Bing and Internet Explorer, for example. But Conversation Translator represents a shift in the way product groups have typically viewed translation, explains Harry Emil, a senior test lead for Microsoft Lync.

From left: Lin Li, senior test engineer for the Lync team and a researcher on language learning; Harry Emil, a senior test lead for Microsoft Lync; and Takako Aikawa, program manager on Microsoft Research’s Machine Translation Team.

“MSR really got us looking at translation as a feature, something that really should be done in line in the product,” he says. “Hopefully we can inspire other teams at Microsoft to view translation this way and start thinking about how to integrate it seamlessly into their products.”

The process doesn’t require much heavy lifting, says Julio Lins, software development engineer in test II. Lins is the author of the original Lync 2010 SDK translation sample that inspired Conversation Translator. “The Microsoft Translator API makes adding translation to your application a simple process,” he says.

Adding translation to its product portfolio could help Microsoft empower individuals across the planet, Emil says. “Imagine kids in Haiti, 15 months after their devastating earthquake, communicating effortlessly with kids in Japan, two months after the disaster there? What could they share? Imagine the possibilities,” he says. “If we have simple tools to break down these language barriers, I think we can find a lot of richness across the planet that would otherwise go unnoticed and untapped.”

A lot of those tools can be found in Microsoft Research’s translation work, says Ross Smith, director Lync client testing. “The translation work was such a natural fit for Lync, and the new platform work in Lync 2010 made it easy for us to work with the Machine Translation team to stitch things together and offer a great solution for customers,” he says. “I’m really glad we have been able to show how a product team can partner with MSR and get something great in front of customers relatively quickly.”

Chris Wendt, principal group program manager of MSR’s Machine Translation Team, says that automatic translation “is a great feature in an interactive communications product like Lync, where the participants can easily recover from any mistranslations and, how Lync nicely shows, can be engaged in submitting a correction, which helps the translation service improve over time.”

Removing language barriers will be the next big step in the evolution of communications, Smith believes. He points out that society has come a long way from the Greek runner at Marathon, the Pony Express, and the rotary dial phone. Today, we can instantly reach almost anyone on the planet. But until now, the language barrier has always been a constant.

“Technology has done wonders for communication, but it has yet to neutralize language differences,” Smith says. “So this is a fascinating opportunity in translation to eliminate the only remaining barrier of instantaneous communications around the world.”

The next step for the team is to make the leap from translation to language learning. They see an opportunity to help Lync users – and users of all Microsoft products – learn new languages in the context of their day-to-day communications to address language and cultural neutrality across all communications, all around the globe.

Related Posts