Microsoft Research Asia has been working with the Chinese Academy of Sciences and Beijing Union University on the project in China, and in the collaboration between engineers, educators, and sign language practitioners has resulted in working demos that are amazing.
The Kinect-based system can capture signs made by a person; recognize the meaning of it using the sign itself as well as body positioning and posture; and then it translates the signs into spoken language or written words. There’s an avatar in the Kinect system that will translate the spoken words of the hearing person into signs for the hearing-impaired person on the other end.
The system is also designed to communicate between Chinese sign language and American Sign Language (ASL), which use different gestures.
Although this is still a research prototype, progress is being made at a rather brisk pace; hopefully this excellent technology will emerge mature and fully-formed soon. Microsoft Research says that there are over 20 million hearing-impaired people in China and about 360 million worldwide.