University of Toronto, Microsoft collaborate in language translator that mimics voice

TORONTO – Consumers have plenty of options to choose from when they want to translate languages, from smart phone apps to Google Translate but a new technology with Canadian origins claims it can translate languages and speak in the same voice as the person talking.

Technology giant Microsoft and experts at the University of Toronto put their brainpower to work for an innovative translation software that listens to a person speak, and can translate the message in the original speaker’s voice, according to a Techvibes report in Canada.

Microsoft’s Chief Research Officer Rick Rashid demoed the product at a presentation in Tianjin, China on Oct. 25.

At the very end of this video below, Rashid tests the product by speaking in English. The translator then changes his message into Chinese and it’s delivered in his voice.

“What happens is we’re basically taking the English text and pushing it through the translation system,” he explains.

Get breaking National news

Get breaking Canada news delivered to your inbox as it happens so you won't miss a trending story.

Sign up for breaking National newsletter Get Started

By providing your email address, you have read and agree to Global News' Terms and Conditions and Privacy Policy.

Rashid concedes the project has some glitches in that word recognition isn’t perfect, but it’s come a long way compared to the accuracy of what’s on the market.

That’s because of deep neural networks technology, which can coherently mimic the way human brains function, opening doors for better speech recognition. With this technology, error rates decrease by over 30 per cent, which Rashid calls a “dramatic change.”

“It’s still not perfect, there’s still a long way to go. I think you can see that we have already made a significant amount of additional progress,” he says.

The translator’s voice sounds identical to the speaker’s because the system uses samples from both the speaker and a native speaker of the translated language.

In a blog post, Rashid says computer scientists have been working relentlessly for the past 60 years in building systems to bridge the gap in language.

Read the full blog post here.

“Most significantly, 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.

Follow @Carmen_Chai on Twitter.