1 paper · 1 filter
Jiaheng Dong, Hong Jia, Soumyajit Chatterjee +3
Speech Foundation Models encounter significant performance degradation when deployed in real-world scenarios involving acoustic domain shifts, such as background noise and speaker…