PhD student position in generative AI for sign language at the School of Electrical Engineering and Computer Science at KTH Royal Institute of Technology.
Project description
This PhD project aims to create cutting-edge models for converting written language to sign language through 3D signing avatars. Using generative AI methods, the research combines large language models with deep probabilistic motion generation in 3D animation. This ambitious project builds on our recent pioneering work in diffusion-based motion generation . Automatic sign language technologies contribute to increased inclusion and have high societal importance, yet the problem is under-studied compared to written and spoken language.
The research takes place at the division of speech, music and hearing, in an interdisciplinary group focusing on modeling and understanding human communication, including speech, language, and gesture using computational and AI-based methods. This is a highly active and dynamic research field spanning machine learning, natural language processing, signal processing, computer vision, computer graphics and perceptual psychology. You will be working alongside other PhD students and faculty working on e.g., natural speech- and gesture generation, dialog systems and social robotics, and more.
The position is financed by WASP, the Wallenberg AI, Autonomous Systems and Software Program and entails participation in WASP doctoral school and doctoral student activities.