AI to decode sign language © Franklin Open
Artificial intelligence is talked about almost daily. If the uses are numerous – and sometimes not very useful – today there is a technological advance that can improve the lives of a large number of people. The use of AI in the medical world is on the rise: a team of researchers at Florida Atlantic University has just developed a model capable of decoding American Sign Language (ASL) in real time. These developments could greatly change the lives and interactions with technology of the hearing-impaired and deaf.
Breaking down communication barriers with AI
Using computer vision technology, researchers have successfully developed an AI model that can transcribe ASL alphabetic signs into text with 98% accuracy. To achieve this result, the researchers trained a model on 29,820 static images of hand gestures. They combined these images with motion tracking technology based on 21 strategically placed signs for sign language.
Training an AI model on American Sign Language © Franklin Open
At this time, the model already achieves remarkable accuracy. The next step for the University of Florida researchers is to extend the model’s training hardware to make it more reliable. The team also says it is working on optimizing the system so it can run on less powerful machines like smartphones. The ultimate goal of this research? Allow for the deployment of a real-time sign language translation tool to bridge communication barriers between the hearing impaired, the deaf, and the rest of the population. Such a system would have many potential uses and we can already imagine how useful it could be for the education or health sectors.
