Microsoft takes voice recognition to a new level

New software synthesizes speech in a foreign language, but in a voice that mimics yours

The day may soon come when not being able to speak in a foreign language may be a thing of the past. Researchers at Microsoft have developed cutting edge software that lets you communicate in a different language. But that’s nothing new you say? Wait, there’s more.

The new software will actually learn the sound of your voice and use it to speak the foreign language. Yes, you heard right. It’s a synthesized version of your own voice that will be heard instead of the standard synthesized computer voice. And the best part is? You don’t even have to learn the other language!

Here’s how it works. The software functions by using speech recognition, followed by translation, followed by the final text-to-speech output in a different language – again, all while retaining the user’s original voice. According to Microsoft it requires about an hour of training time to learn the nuances of your voice and work – a significantly short period compared to the amount of time it would take to learn a new language. Once released the technology should go a long way in making language tutoring software more personal.

Presently the program can translate between any pair of 26 different languages, including English, Mandarin Chinese, Spanish and Italian. Researchers at Microsoft's research lab in Beijing are also working on adding speech recognition capabilities - much like the conversation mode for Google Translate - which would make it an essential tool for people who travel to many different countries, and who don’t have the time to learn how to communicate with the locals.

For now there’s no official word out of Redmond as to when this technology will be released to the public. But it’s sure to make a splash when it does. So stay tuned…

Alex Dupont

See Also