Research projects

DESIGN project aims to develop a mathematical representation of Spanish Sign Language (LSE) to improve its automatic understanding and translation using artificial intelligence techniques. The project will develop advanced models capable of interpreting the meaning of signed messages within their context by integrating Large Language Models (LLMs) and Vision-Language Models (VLMs). The resulting technologies will be validated in three use cases: translation in patient–physician interactions, medical appointment dialogue management, and Spanish Sign Language learning assessment. In addition, DESIGN will advance automatic annotation techniques and develop a recognizer for at least 2,000 signs, providing a foundation for future sign language recognition and translation systems.