Mad or sarcastic? Your computer might know someday

Siri may be able to recognize your voice, but she can’t really pick up nuances like sarcasm or outrage yet. But researchers are working on a way to allow computer software to recognize subtle cues. Valerie Freeman, a Ph.D. candidate in the Department of Linguistics at the University of Washington (UW) and colleagues are working on a project for the Automatic Tagging and Recognition of Stance (ATAROS) which aims to train computers to recognize the various stances, opinions and attitudes that can be revealed by human speech. The researchers recorded the voices of 68 people of different ages and backgrounds listening for cues that could reveal the speaker’s stance on a topic.

What is it about the way we talk that makes our attitude clear while speaking the words, but not necessarily when we type the same thing? How do people manage to send different messages while using the same words?

Valerie Freeman, a Ph.D. candidate in the Department of Linguistics at the University of Washington

The researchers found that when pairs of people worked together on various tasks, they tended to speak faster, louder and with “more exaggerated pitches” when they had a strong opinion than when they had a weak one. For example, people talked more quickly when they worked on more engaging tasks, such as balancing an imaginary budget, as opposed to less complicated tasks like arranging items within an imaginary store, Freeman said. In addition, when people were engaged in a subject, they spoke less fluently, the researchers found. They tended to speak in false starts, repeat themselves and say “um” more often than when they were less interested in the topic of conversation.

Projects like ATAROS are working to help computers learn how to figure out what people really mean when they speak, so that in the future computers will be capable of responding in a much more ‘human-like’ manner.

Valerie Freeman

Technology