LIS2Speech. Increasing accessibility for deaf people: opportunities and challenges of artificial intelligence

The limited knowledge of LIS restricts the ability of deaf individuals to effectively communicate with the rest of the population. This is critical in contexts such as public administration, healthcare, or tourism. By combining deep learning and computer vision, it is possible to develop smartphone apps and AR goggles capable of translating LIS in real time. We will share scientific findings and some proposals to address the main challenges: the availability of gloss datasets and video stream processing