LTI Professor Alan Black was featured in two recent articles about speech recognition technology.
Alan Black, a Scottish computer scientist and professor in Carnegie Mellon University's Language Technologies Institute, was recently featured in two articles discussing developments in voice-activated technology.
The New York Times article "Creating a Computer Voice That People Like" discusses the roots of modern speech synthesis technology, which Black’s work investigates. Black said that even with major progress, speech synthesis systems have not yet achieved humanlike perfection because users still can't control how they deliver a command to a specific system and how it, in return, analyzes it.
Black elaborated on this phenomenon in an article in The Guardian, where he explained that this issue arises because humans tend to have a "machine voice" that encourages them to be more polite and to drop colloquialisms. For example, people will actually change how they speak with their families when they're on the telephone or talking to a machine. Black believes that in the future, programs like Siri will become more familiar with the user, and will understand language patterns like a close friend.
"Dialogue systems at the moment work pretty well," Black said. "I think what’s probably going to happen is a much more long-term rapport. It will be able to answer the question sort of before you ask it."
Read more about Black’s work with speech and voice recognition technology on The Guardian and The New York Times. His conversation with The Guardian was also quoted in an article on How Stuff Works.