Picture this, an English-speaking man carries on a full conversation with a German-speaking women. Did I mention that the man doesn’t understand a word of German and the woman doesn’t understand a word of English? Yet, they still understand each other.
On stage at the Code Conference in Rancho Palos Verdes, Calif., Gurdeep Pall, corporate vice president of Skype and Lync at Microsoft, demonstrated Skype Translator, a new app that translates multilingual voice calls in real time. This means that users can speak in their native tongue to someone who speaks a different language while Microsoft translates. By the end of this year, Skype Translator will be available as a Windows 8 beta app.
“Imagine in the very near future technology allowing humans to bridge geographic and language boundaries to connect mind to mind and heart to heart in ways never before possible,” Pall writes in his blog post.
In the lively demo, Pall uses Skype to initiate a video and voice call with his German-speaking demonstrator, Diana. As they converse in their own respective languages, subtitles in both German and English appear at the bottom of the screen accompanied by an almost real-time audio translation. And, get this; the English translation of Diana speaking German even comes with a female voice.
According to Satya Nadella, the new CEO of Microsoft, the company's machine translation group has been working on translation, speech recognition, and speech synthesis for the past 15 years.
Skype Translator fuses “Skype voice and IM technologies with Microsoft Translator, and neural networking-based speech recognition,” Pall writes in his blog post. Although this sounds like a mouthful, this technology stems from transfer learning. It's basically the computer version of "practice, practice, practice."
“It’s brain-like in the sense of its capability to learn,” Nadella said at the Code Conference this week. But, regarding the nitty-gritty of how exactly the technology works “quite frankly, none of us know quite exactly why,” he says, and calls it “magical.”
Theresa Chong is a video host and multimedia technology journalist based in Palo Alto, Calif. As on-camera talent, she has performed science experiments for “Discovery News,” explained how virtual reality works for USA Today, and interviewed Adam Savage for IEEE Spectrum. She has written about wearables for Scientific American and travel tech for Architectural Digest. With a DSLR, GoPro, and green screen by her side, she has produced digital videos of robots, driverless cars, and 3D printing. She earned a master’s degree from Northwestern University’s Medill School of Journalism, and in a prior life she worked as a civil engineer.