Microsoft has launched a language translation tool “patterned after
the human brain” that can instantaneously translate spoken language.
The translation software can take a user’s spoken English word and translate it into Mandarin in the user’s own voice.
Microsoft chief research officer Rick Rashid first unveiled the
breakthrough technology during the company’s Research Asia’s 21st
Century Computing event in China last month.
“We have attained an important goal by enabling an English speaker
like me to present in Chinese in his or her own voice, which is what I
demonstrated in China,” Rashid wrote in a blog post.
“It required a text-to-speech system that Microsoft researchers built
using a few hours speech of a native Chinese speaker and properties of
my own voice taken from about one hour of pre-recorded (English) data,
in this case recordings of previous speeches I’d made.
“Though it was a limited test, the effect was dramatic, and the
audience came alive in response. When I spoke in English, the system
automatically combined all the underlying technologies to deliver a
robust speech-to-speech experience—my voice speaking Chinese.”
The translation software is based on a technique dubbed Deep Neural
Networks (DNN). Instead of using the “hidden Markov modeling” technique,
which centers translation on training data from a number of speakers,
DNN uses human brain behavior to provide superior speech recognizers.
With DNN, Microsoft has cut translation error by more than 30
percent, compared to the Markov method, Rashid wrote. Older versions
made errors once in every four or five words. DNN, however, makes errors
once in every seven or eight words.
To make Rashid’s Mandarin speech possible, a few hours of speech by a
native Chinese speaker was recorded for the software to use. Microsoft
then recorded about an hour of Rashid’s voice and fused the two voices
together.
Rashid, who calls Microsoft’s technology “very promising,” believes
the company will have systems that can completely break down language
barriers in only a few years.
“In other words, we may not have to wait until the 22nd
Century for a usable equivalent of Star Trek’s universal translator, and
we can also hope that as barriers to understanding language are
removed, barriers to understanding each other might also be removed,” he
wrote.
A video of Rashid’s English speech being translated into Mandarin can be viewed here.
Post from: SiteProNews
