Microsoft may boldly go where no one has gone before.
At a demonstration in China on Oct. 25, Microsoft's chief research officer Rick Rashid revealed a new technology seemingly straight out of "Star Trek." In the classic sci-fi television show, a gadget known as the "universal translator" (as seen here) was used to decipher languages in the 22nd century. And now, Microsoft has created a program that takes spoken English and translates it to spoken Chinese in real time.
Pretty futuristic, right?
"In the realm of natural user interfaces, the single most important one -- yet also one of the most difficult for computers -- is that of human speech," Rashid wrote on the company's blog.
Let's let the translator speak (pun intended) for itself. Watch below to hear Rashid's speech translated into spoken Chinese on stage (the fun starts around the 6-minute mark):
Speech recognition has made achingly slow progress, as Rashid tells it. First, speech recognition was done by matching waves from spoken sounds with those known to be associated with certain words. In the 1970s, scientists built statistical speech models by analyzing many speakers' voices.
In 2010, Microsoft made another breakthrough, it says, by analyzing speech with computer programs that patterned themselves on human behavior. Microsoft says it was able to "reduce the word error rate for speech by over 30 percent compared to previous methods."
The MIT Technology Review notes that the tech giant has previously demonstrated "synthesized speech to match a person's voice," but the prototype was only able to speak typed text. This latest translator quickly understands spoken words (which is a feat in itself), while also picking up on the cadence of a user's voice.
"In other words, we may not have to wait until the 22nd century for a usable equivalent of 'Star Trek’s universal translator," Rashid said. "[A]nd we can also hope that as barriers to understanding language are removed, barriers to understanding each other might also be removed."
Of course, Microsoft has plenty of competition in the realm of voice recognition software. Apple has produced Siri, the iPhone's voice-controlled assistant, while Google has created the Google Search voice app, dubbed the "Siri Killer." But both consumer products are still rife with errors, leading some customers to mock these services via Tumblr and snarky blogs.