Project Aslan Creates 3D Printed Sign Language Robot

[Clare S]

Three years ago, three students from the University of Antwerp started Project Aslan, which involved 3D printing a robot capable of translating text into sign language. The project was begun in order to address the global communication barrier between hearing people and hearing-impaired people, and will be open source once it is complete. Thanks to a partnership with 3D Hubs, the robot will be available to more than 140 countries around the world. Read more at 3DPrint.com: http://3dprint.com/184677/3d-hubs-project-aslan/

[Penelope]

Clare S,
There are many projects like this that become problems when people who are not fluent in sign language, have limited experience with deaf people, and are unfamiliar with how sign language works beyond simple phrases and fingerspelling are the ones who undertake these projects. First of all, deaf people prefer not to referred to as "hearing impaired", as they do not consider themselves impaired.
References:
http://www.deaflinx.com/DeafCommunity/identity.html
http://www.washington.edu/doit/how-a...typically-used

Second of all, a robot that simply translates a spoken language into signs, word for word and letter for letter, is the exact same thing as simply typing up a transcript of that spoken language. Why not just have a CART typist in the room transcribing everything? The technology already exists for this, nothing new needs to be invented, tweaked, or more money spent. The problem isn't that deaf people can't read written language, of course they can, they know which signed letter corresponds to which written letter of their alphabet. The problem comes when, because not all deaf people are fluent in both their written and signed language, they would know sign language than their country's spoken language and while they know the basic written language, might not know more complex words or the MEANING of the overall written or robotically translated message. For example, in court if the judge asked the deaf person "Why did you run from the police?" the robot would have to be capable of determining how to sign the word "run". Sign language is CONCEPTUAL, so the robot might sign run as in "jog" or run as in "the car motor ran" or the "the water ran" or "the boss ran the meeting", each concept has a completely different sign to match the varying concepts. A human interpreter automatically process which "run" the judge is talking about and can sign it correctly. A robot cannot. If a robot signed to you "Why did you control the police?" because it decided you were talking about "running a meeting", in which case the sign for "run" could also be representing the concept of control, managing, etc, it could get very confusing.

This idea is similar to this one:
http://www.huffingtonpost.com/entry/...b0f309baeee06d

And from the perspective of people who actually use sign language, it just doesn't work. It would be great if people working on these projects would actually bring in deaf people and sign language experts to consult, although perhaps they wouldn't want to be told that their idea simply isn't a good one.

The danger here is that a judge or police station might say "wow this is a great idea and will save us lot's of money!" and they have no clue about how sign language actually works or how this type of technology could lead to huge misunderstandings and confusion.

A better alternative when an in person interpreter isn't available, is VRI, video remote interpreter, where the interpreter is on the screen. At least you still have a human brain involved that can determine situation, context, concept, and will successfully interpret and and not just translate.

[endure]

At least Project Aslan is trying to do something. Apart from criticising other people's efforts what are you doing?