University of Toronto graduate students George Dahl and Abdel-rahman Mohamed have helped develop a new translation algorithm that could enable technology that translates what you say into another language — in your own voice.
Dahl and Mohamed, both Ph.D students in computer science, first developed the acoustic modelling algorithm in 2009. After it set new benchmarks in computerized translation, Microsoft came calling and offered them internships to work on the technology.
“It gets rid of the communication barrier that exists,” said Mohamed. “I could learn English and not have to go to school to do it.”
Uniquely, for such a groundbreaking advance, Dahl and Mohamed opted against patenting the new technology, allowing companies like Google and IBM to apply it to their own speech-to-text programs.
Dahl said he expected the algorithm to find future use in computer vision, computer-aided drug discovery, and natural language processing.
With files from the Toronto Star.
