Speak 26 languages with Microsoft's new translation software

Researchers at Microsoft have developed a tool that can translate your speech into another language while still maintaining the sound of your voice.
Written by Sarah Korones, Contributor

New software from Microsoft Research might make it possible for speakers of different languages to converse effortlessly.

Researchers at the tech company have developed a tool that can translate your speech into another language—in a voice that actually sounds like your own. The system could be used to improve tools for travelers or to make the process of learning a language a bit more personal.

The tool’s creators demonstrated the software during a conference at Microsoft’s headquarters in Redmond, Washington and explained that the system starts with speech recognition. From there, the speech is translated into text and followed by a text-to-speech output in a different language. But rather than a cold, computerized voice, the final output should sound remarkably like the original speaker.

Technology Review 's Tom Simonite explains:

The system needs around an hour of training to develop a model able to read out any text in a person's own voice. That model is converted into one able to read out text in another language by comparing it with a stock text-to-speech model for the target language. Individual sounds used by the first model to build up words using a person's voice in his or her own language are carefully tweaked to give the new text-to-speech model a full ability to sound out phrases in the second language.

Developers hope the system will help students learning a new language, saying that hearing foreign phrases in a person’s own voice could be easier to imitate. The tool’s creators also believe their software can improve navigational phone apps, translating signs and directions with ease.

The software can translate between any pair of 26 languages, including Mandarin Chinese, Spanish, and Italian.

The full audio demos are available at Technology Review.

Image: Alexandre Vialle/Flickr

This post was originally published on Smartplanet.com

Editorial standards