Barriers are broken when you talk! And soon you’ll be able to communicate with a person who can’t speak or understand your language. Mobile carrier AT&T is carrying out trials that let customers send and receive text messages that are automatically translated from one language to another and back. All you have to do is to register your phone number with a preferred language. When a text message or SMS is sent to that number in a different language, the text message is automatically translated before being relayed to the person’s phone. The service could be particularly helpful for businesspeople who work with contacts overseas.
Mazin Gilbert, vice president for intelligent systems research at AT&T Labs demonstrated the revolutionary automatic translation of text messages at the AT&T Foundry inPalo Alto,California. He said that the technology works on any cell phone and doesn’t require any software to be added to a handset nor it requires customers to hop from an existing service.
“If my friend sends me a message in Spanish, then AT&T recognizes the language and translates it to English. Then I can send back in English, and it is translated into Spanish for her. We can go back and forth and have a conversation – we’re trying to break down language barriers between people.”
Currently, the technology supports translation between English and Spanish, but AT&T has already deployed technology capable of handling six more languages in existing products, says Gilbert, so they could easily be added to the text message translation system.
When asked how to signal to a person that a message they received has been translated correctly. Gilbert replied:
“The technology is not always perfect,” acknowledges Gilbert, and could introduce errors or confusing phrases to some messages.
Earlier this year in April, AT&T demoed a similar technology that allowed voice to be automatically translated between Spanish and English, without noticeable delay. AT&T lab then demonstrated a real-time conversation between English and a Spanish speaker. Spoken words were rendered into text, translated, and spoken by machine to the listener on either end.