Skype Translator may be the biggest advancement in real-time language translation since the babelfish on “Hitchhiker's Guide to the Galaxy.”
The app, currently in a limited trial for Windows 8.1 and for English and Spanish speakers only, listens to what you say on a Skype video or audio call and then translates immediately. It can also translate the text you type during a Skype instant messaging chat. The chat works for over 40 languages beyond Spanish including German, French, Arabic and even Klingon.
Recently, I had a chance to test out the preview release by calling a Spanish-speaking person on the Microsoft team named Ignacio Perez Lozano, who lives in Spain. A calm and collected millennial, he was game to let me try a few tougher words and local idioms.
The technology that makes this happen involves a third-party: a robot. When you place a call, a translation engine serves as an interpreter. It literally joins the call and processes what you say.
According to Vikram Dendi from Microsoft Research, the back-end processing comes from many lessons the tech giant (that now owns Skype) has learned about natural language processing over the past 15 years. For now, they only offer Spanish and English, but will expand soon.
“This approach allows for a better user experience, as we will not release languages until they meet a certain standard of accuracy, including input from our internal users and partners,” he told me. “We’ll be listening to feedback to determine what platforms are most important to customers, which languages they want to use, and how they plan to use Skype Translator.”
The big surprise here is that Skype Translator really does work almost instantly. As I spoke, the robot sometimes started translating mid-sentence. Only in a few cases did my words not translate quite accurately. I asked about the weather in Spain -- turns out, it is sunny and mild right now -- and asked a few easy questions about working in a foreign country.
In a few cases, I tried to find out if the translation could understand colloquialisms and slang, and if it could keep up with a few well-placed words most people would not understand.
I first said "I just lap-tested a Bugatti in Miami and it was sick." My helper in Spain could see the real-time text of the translation in both Spanish and English, and he understands both languages so he knew I meant I had driven on a track and that I felt the car was amazing (not broken or at the shop). However, the exact translation used the word "enfermo," which means ill.
Next, I tried saying words like "egregious:" and "serendipitously," which are not part of everyday vocabulary. Egregious means "unusually terrible" and translated correctly by both voice and instant message. Serendipitously means "out of the blue" and the word did not translate quite right (it said “serendipitous Lee”) and Lozano didn’t know the meaning in Spanish.
Since I'm from Minnesota, I also tried the ultimate test. I mentioned something about going to a potluck on Sunday. Skype Translator used the English word, but Lozano didn’t know it. That said, even people fluent in English from other countries have a hard time with local idioms. And, when I explained it meant everyone brings a meal to share, he understood easily.
Skype Translator works fast, it's reliable, and it understood even unusual and complex multi-syllable words. The main takeaway is that a computer is more than capable of translating what we say. It has a hard time relaying what we mean in every context. For that, Microsoft researchers might need to teach it more than just the words but the context of what we are really trying to say. It would sure help with any globe-spanning misunderstandings.