Cell phones have just gotten more accessible now that people can communicate on them via sign language. A group at the University of Washington has developed software to make this happen. They recently received a National Science Foundation grant for a 20-person field project that will begin next year in Seattle and hope to get the prototype on the market in one to two years.

The main barrier to real-time two way video, according to the UW press release, is that the US hasn't had fast enough connections. As I reported in 2007, the considerably faster networks overseas -- called 3G -- can already support videophones. 3G is available in parts of the US but not everywhere.

The team's breakthrough was to sacrifice the background so video can fit on the current US network. They call this MobileASL. This decision was based on eye-tracking experiments, which showed that people spend the most time looking at a person's face while they're signing. The team then figured out how to keep hands and faces at high resolution but cut back video quality for the background. They are now looking into saving battery power by not using high quality video of hands and faces when people aren't moving their hands.

The neat thing about this is that MobileASL may work for us lipreaders too. It was hard to determine lipreadability because the people on the sample videos are using ASL; they're not moving their lips much, if at all. But the possibility is there. When I brought this up to Eve Riskin, the team's leader and a UW professor of electrical engineering, she admitted they hadn't thought about this aspect, but she plans to follow up about it with her collaborators.

For now, both parties have to have the same software, said Riskin. "However, since the encoder we are using is based on a video compression standard (H.264), our video should be decodable by a H.264-compliant decoder. But right now, we don't have access to such a system," she said.

When asked what the largest screen size is, Riskin said the group hasn't really discussed this. They should be able to work with any size screen on a cell phone simply by interpolating or down sampling the video.

The idea is that the system can be used on the existing cell phone network, since not everyone can afford 3G. "It isn't fair that people who are deaf or hearing impaired should have to pay more for their services than hearing people," said Riskin, who hadn't heard of a TAP plan.

UW reports that a major cellular network provider has expressed interest in the project. In the meantime, the team will be running the 20-person field test: measuring satisfaction and how much the participants use the technology. Those of us outside Seattle will just have to wait and see how this develops.

© Copyrighted material, used by permission. This article can not be copied, reproduced, or redistributed without the express written consent of the author. Author's views not necessarily those of i711.com.

View Comments  |  Add Comment  |   Email this Article