🏆 Won 3rd Place at HackHershey 2019
The ASL (American Sign Language) Translator uses machine learning to accurately translate finger spelling to text. It also includes various extra gestures, like space and delete, to make a more complete for finger spelling experience. In the future we hope to increase the accuracy, as well as to expand beyond fingerspelling to the entire language of ASL.
This system uses a special type of AI, known as machine learning, to recognize different hand shapes. This is known as a Convolutional Neural Network, which is the state of the art for image classification and recognition. This model can be retrained on many more hand shapes, which means that in the future, we can train this model to recognize much more of ASL.