acm-header
Sign In

Communications of the ACM

ACM TechNews

Big Brother Untangles Baby Babble


View as: Print Mobile App Share:
Professor Deb Roy of Media Lab's Cognitive Machines group

"We wanted to understand . . . how the interplay of innate and environmental influence makes us who we are and how we learn to communicate," says Professor Deb Roy of MIT Media Labs.

Deb Roy, an artificial intelligence researcher at the Massachusetts Institute of Technology Media Lab, is exploring children's development of communication skills through an unusual research strategy called the Human Speechome Project. He equipped his residence with a series of cameras, microphones, and terabytes of storage with which he could record his and other people's interactions with his infant son. To mine meaningful data from the massive amount of information collected, the project team devised a series of software tools.

The Total Recall tool enables researchers to rapidly scan through any part of the data, while the Blitzscribe software pinpoints speech in the recordings and deconstructs it into easily transcribed sound bites. Meanwhile, the TrackMarks human-computer system analyzes video footage and provides information such as where people are in relation to one another and the position of their heads.

"Just as the Human Genome Project illuminates the innate genetic code that shapes us, the Speechome Project is an important first step toward creating a map of how the environment shapes human development and learning," says Media Lab director Frank Moss.

Realizing that his conclusions about children's communications development may be difficult for the scientific community to accept because they are based on analysis of only one child and are unlikely to be replicated, Roy has developed an easily installed device called the Speechome recorder. The device features an overhead microphone and camera, a touch-screen display, and sufficient storage to retain several months of recordings. Roy's ultimate goal is to develop a machine imbued with human-like learning capability.

From BBC News
View Full Article

 

Abstracts Copyright © 2009 Information Inc., Bethesda, Maryland, USA


 

No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account