Ongoing Projects
Active research programmes with open questions and upcoming results.
SignGPT is a UKRI EPSRC Programme Grant focused on the development of an AI-powered translation system capable of unconstrained, bidirectional translation between British Sign Language (BSL) and English. The project will build the first generative predictive transformer for sign language, combining computer vision, sign linguistics, and machine learning. It is led by an interdisciplinary team from the University of Surrey, University of Oxford, and University College London, with direct involvement from Deaf organisations and community partners.
Grant funding and additional support from Google's philanthropic arm, Google.org, to develop artificial intelligence research paving the way for instant sign language translation. Working with Signapse as a delivery partner, the project translates key websites into sign language, boosting digital inclusion for the 600,000 Deaf people in the US and UK for whom sign language is their first language.
Understanding Multilingual Communication Spaces (UMCS) is a £3.5 million UK-Japan research project developing human-centred AI and augmented reality systems for real-time translation across British Sign Language, Japanese Sign Language, English and Japanese. The project focuses on authentic Deaf-to-Deaf conversations, capturing natural interaction patterns such as turn-taking, backchannels, repair strategies and shared visual attention, rather than relying only on interpreter-facing video data.
Prior Projects
Projects that have concluded, with released datasets, code, and papers.
The goal of the IICT Flagship is to develop information and communication technologies (ICT) for persons with disabilities. The flagship targets five applications in the context of accessibility: text simplification, sign language translation, sign language assessment, audio description, and spoken subtitles. Each application constitutes its own subproject, with tight connections through shared AI technologies. The University of Surrey is included due to its extensive expertise in AI for sign languages.
The goal of SMILE is to pioneer an assessment system for Swiss German Sign Language (Deutschschweizerische Gebärdensprache, DSGS) using automatic sign language recognition technology. The project develops robust recognition models tailored to DSGS and integrates them into a learner assessment framework to support Deaf education and sign language acquisition.
This project tackled the challenge of automatically learning to recognise dynamic activity in broadcast footage, with demonstration activities in both sign language and more general actions and activity. The work developed methods for weakly supervised activity recognition from large-scale broadcast data, bridging the gap between constrained lab settings and real-world video.
Making Sense focused on improving investigative capability through the application of science and technology. The project explored how advanced computer vision and machine learning techniques can be applied to support investigative processes, including the analysis of visual content at scale.
DictaSign aimed to enable communication between Deaf individuals by promoting the development of natural human computer interfaces (HCI) for Deaf users. It researched and developed recognition and synthesis systems for sign languages at a level of detail necessary for recognising and generating authentic signing. Research outcomes were integrated into three prototypes: a Search-by-Example tool, a Sign Language-to-Sign Language translator, and a sign Wiki — enabling knowledge-sharing within and across Deaf communities.