Egyptian and Nigerian English join the Open ASR Leaderboard

The Open ASR Leaderboard is expanding its evaluation datasets by adding private data for Egyptian and Nigerian English accents. This collaboration with Liva AI introduces new conversational speech datasets to improve representation and trust in automatic speech recognition evaluations. The initiative aims to provide more robust, held-out test sets to minimize the risk of models overfitting to public benchmarks. As part of this update, the leaderboard will now present a more detailed breakdown of model performance across different accent categories based on Kachru’s Three Circles of English.