The Inquirer-Home

Microsoft demonstrates live speech translation

English to Chinese on the fly
Fri Nov 09 2012, 16:39
Microsoft logo

SOFTWARE SHOP Microsoft has shown off speech translation technology that turns English into Chinese on the fly.

Microsoft chief research officer Rick Rashid showed off the technology in a video, and it does look pretty good. Speech was translated as it happened, though we can't speak Chinese so we don't know if it made any mistakes.

Rashid revisited the speech in a blog post where he discussed how automatic machine translation has matured.

He said that the demo is the result of 60 years of work, not just by Microsoft, and some breakthroughs in the last two years. These breakthroughs, he said, have helped take the software away from the error rates of 20-25 percent that it previously suffered.

"Just over two years ago, researchers at Microsoft Research and the University of Toronto made another breakthrough. By using a technique called Deep Neural Networks, which is patterned after human brain behavior, researchers were able to train more discriminative and better speech recognizers than previous methods," he said.

"We have been able to reduce the word error rate for speech by over 30 percenty compared to previous methods. This means that rather than having one word in [four] or [five] incorrect, now the error rate is one word in [seven] or [eight]."

In the video demo Rashid's speech is translated into Chinese while still retaining his style of speaking. This makes the experience more personal and natural, and takes it even further away from other systems.

He said that the software turns his English to Chinese in two stages.

"The first takes my words and finds the Chinese equivalents, and while non-trivial, this is the easy part. The second reorders the words to be appropriate for Chinese, an important step for correct translation between languages," he explained.

"Of course, there are still likely to be errors in both the English text and the translation into Chinese, and the results can sometimes be humorous. Still, the technology has developed to be quite useful. 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."

It's in the video, it happens around two minutes from the end. It is pretty impressive, and the Chinese audience gives it the applause that apparently it deserves.

"Though it was a limited test, the effect was dramatic, and the audience came alive in response," he said.

"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 results are still not perfect, and there is still much work to be done, but the technology is very promising, and we hope that in a few years we will have systems that can completely break down language barriers." µ


Share this:

blog comments powered by Disqus
Subscribe to INQ newsletters

Sign up for INQbot – a weekly roundup of the best from the INQ

INQ Poll

Happy new year!

What tech are you most looking forward to in 2015