American Software Meant for the Deaf Can Help All

Mobil ASL, the software that the University of Washington has just developed for deaf and hard-of-hearing Americans may be the first software that enables an American to use sign language over mobile phones. But thanks to its unique features, this software may also be ideal for the developing world as well.

Discussing the significance of this development Eve Riskin, the principal investigator, who is a professor at the Washington University and a part of the team that developed this software explained, "The software is tailored for American Sign Language. But since it is based on cutting edge technology, it could be modified for use for any sign language because all sign languages basically have similar gestures."

According to Prof. Riskin, the biggest challenge of transmitting sign language -- which the deaf use to communicate with each other -- over cell phones is not to compress the video of the gestures of sign language compact enough to be transmitted and received over cell phones, but rather to use an algorithm simple enough for a cell phone to do the real time encoding and decoding fast enough.

"Cell phones posses far lower hardware resources in terms of processor speed and memory, and that is a big hurdle for video communication over cell phones," she said.

Riskin added that the advantage of Mobile ASL lies in a sign language encoder designed by the team that not only adheres to the ITU compression standard -- new H.264/AVC -- but also nearly doubles the compression ratio. Consequently, the software can compress the gestures tight enough to make it comprehensible on low-resolution video.

"This is why we feel that the new software would a big help for people in the developing world because all developing countries suffer from poor bandwidth problems" says Riskin. "And even if some of them do have access to faster networks, they are usually very highly priced for most."

Which isn't to suggest that Mobile ASL is not a significant development for Americans as well. "A lot of people are excited about this," said Riskin. That's because, even as faster networks are becoming more common in the United States, such networks are still not fast enough for high rates of data transmission, and that means that there is still a need for software that can operate on slower systems.

Moreover, faster networks are not available everywhere and they also cost more. "We don't think it's fair for someone who's deaf to have to pay more for his or her cell phone than someone who has hearing," feels the team.

Of course, cell phones often have become indispensable for the deaf, just as they are for others - even if text messaging was the communication underpinning for the deaf. That, according to many deaf users, was very limiting, to say the least. For one, text messaging doesn't allow one to communicate rapidly one's native language. And  more importantly, it is slow and does not convey emotions at all.

"Video is much better than text-messaging," says Jessica DeWitt, a UW undergraduate in psychology who is deaf and is a collaborator on the Mobile ASL project. However, low data transmission rates on U.S. cellular networks, combined with limited processing power on mobile devices, have so far prevented real-time video transmission with enough frames per second that it could be used to transmit sign language.

So, that brings us to the next question; is the software ready to hit the market? Not just yet. 

 "It is still in its research stage and not ready for prime time. Work is still on to make it user friendly. But after that we would like a cell phone company to pick it up and release it over the cell phone system," says Riskin.

"The team is already in discussion with a major cellular network provider that has expressed interest in the project," she said, adding that it may hit the market in about 18 months.

And what if the talk fails? "No worries," says Riskin. "We will throw it on the web and turn it open source for anyone to modify it for prime time use."

For that matter, the team is also working to a future version that will incorporate custom tools to get better quality and a feature that will identify when people are moving their hands, to reduce battery consumption and processing power when the user is not signing.