September 14, 2009

Computers better lip-readers than humans

British scientists say their finding that computers are better lip-readers than humans may lead to improved lip-reading training for the deaf.

A new study by the University of East Anglia compared the performance of a machine-based lip-reading system with that of 19 human lip-readers. The researchers found the automated system significantly outperformed the human lip-readers -- scoring a recognition rate of 80 percent, compared with only 32 percent for human viewers on the same task.

The scientists also determined the computers were able to exploit very simplistic features that represent only the shape of the face, whereas human lip-readers require full video of people speaking.

This pilot study is the first time an automated lip-reading system has been benchmarked against human lip-readers and the results are perhaps surprising, said the study's lead author Sarah Hilder. With just four hours of training it helped them improve their lip-reading skills markedly. We hope this research will represent a real technological advance for the deaf community.

The findings by Hilder, Richard Harvey and Barry-John Theobald appear in the Proceedings of the International Conference on Auditory-Visual Speech Processing. The findings were presented during the conference Saturday at the University of East Anglia.