Hogan Lovells Social Impact Practice HL BaSE announces partnership with the Centre for Digital Language Inclusion to advance inclusive AI for impaired speech as part of its advice to the Global Disability Innovation Hub

Last month, CDLI unveiled its first comprehensive collection of impaired speech datasets on the open data platform Hugging Face. The launch represents an important step in addressing the longstanding underrepresentation of African languages and dialects in artificial intelligence and natural language processing technologies.

Announced at Africa Tech Summit Nairobi, one of the continent’s most influential technology conferences, the release brings together datasets capturing Kenyan English, Swahili, Ugandan English, and Luganda. The collection represents the first African dataset specifically focused on impaired speech variations, reflecting the natural variations, code-switching, and dialectal features that characterise everyday communication across East Africa.

By making these datasets openly available, CDLI aims to enable researchers, developers, and organisations to build more inclusive voice and language technologies. In particular, the data will support the development of applications that allow people with speech differences and disabilities to communicate more effectively through digital tools.

The HL BaSE team is advising the Global Disability Innovation Hub, which hosts CDLI. The firm’s involvement reflects its ongoing commitment to supporting social innovators and organisations using technology to address systemic barriers and expand access and opportunity.

The initiative is led by UCL’s Global Disability Innovation Hub and supported by Google.org and UK International Development through the AT2030 Programme. CDLI works with a range of local and international collaborators, including Modal, The Research Center Trustworthy Data Science and Security, Talking_Tipps Africa, Senses Hub, the University of Ghana, Strathmore University, and Hogan Lovells’ HL BaSE practice.

Yasmin Waljee, International Pro Bono Partner at Hogan Lovells, said:

“Technology should work for everyone, yet many AI systems today fail to recognise the full diversity of how people speak. We are proud to support the Centre for Digital Language Inclusion and the Global Disability Innovation Hub in their work to make voice technology more accessible and inclusive. By helping ensure that speech differences and underrepresented languages are reflected in the data that powers AI, this initiative has the potential to open up digital tools and opportunities for many people who have historically been left out.”

Dr Katrin Tomanek, AI Tech Lead at CDLI, said:

“This is a huge milestone for our team and for African language technology. By publishing these datasets, we’re addressing a critical gap in AI development and ensuring that the rich linguistic diversity of East Africa, including those who have speech differences, is recognised and accessible to researchers and developers worldwide.”

The datasets are now publicly available open-source on Hugging Face with an access protocol designed to ensure responsible use while maintaining openness for legitimate research and development purposes.

For more information, visit: www.cdl-inclusion.com.