Sheffield academics lead landmark AI study to help systems understand regional accents

A groundbreaking study led by researchers from the University of Sheffield is set to improve AI’s capacity to understand UK regional accents and dialects within the public sector.

The pioneering study is being led by Dr Chris Montgomery from the University of Sheffield’s School of English in collaboration with ICS.AI, one of the UK’s fastest-growing artificial intelligence companies.

The main aim of the study is to improve the efficiency of conversational AI digital systems in the public service sector when handling callers with different regional accents. This includes technology used by bodies such as local councils to assist users to reach the correct information channels when making an enquiry.

It is hoped the study will ensure the development and innovation of conversational AI in public service scenarios will encompass the diverse, wide-ranging accents found in the UK.

This follows a recent survey which revealed 52 per cent of UK residents are concerned AI may struggle to understand accents or dialects. In Wales, this rises to 57 per cent and increases again in Northern Ireland in Scotland to 67 and 71 per cent respectively.

Dr Chris Montgomery, Senior Lecturer in Dialectology at the University of Sheffield, is advocating for the application of sociolinguistic theory onto the assessment of public sector conversational AI.

He said: “As AI is frequently used to direct calls across public services, we need to ensure the technology can understand the range of accents and dialects it may be faced with.

“Similar to when you learn a new language, you are often taught about how that language can vary throughout the regions where it is spoken, so we need to apply that same approach when developing AI.”

via Wikimedia Commons

Dr Montgomery recently completed a systematic review of five decades worth of peer-reviewed research on accent and dialect variation across Great Britain which forms the foundation of the project.

Dr Crispin Bloomfield, Chief Education Solutions Officer at ICS.AI, says public sector AI tools must be accessible to everyone regardless of their accent.

He said: “This collaboration empowers ICS.AI to apply established sociolinguistic evidence directly to how conversational AI is evaluated in live public service environments, helping us to build inclusivity in a transparent and scientifically grounded way.

 “The collaboration reflects the University of Sheffield’s commitment to independent thinking and a shared ambition, demonstrating how creative minds at Sheffield are shaping solutions to national and global challenges.”

The University of Sheffield will continue its research into misrepresented dialects to build its data to support future work with ICS.AI.

For more of the latest news, guides, gossip and memes, follow The Sheffield Tab on Instagram, TikTok, and Facebook. 

Featured image via Wikimedia Commons