1 paper
Mustafa Ozan Duman, Ahmet Emir Dirik
Automatic Mean Opinion Score (MOS) prediction is essential for evaluating large-scale synthetic speech and audio enhancement systems, yet models frequently struggle with domain shi…