1 paper
Xuanru Zhou, Yiwen Shao, Wei-Cheng Tseng +1
Current audio pre-training seeks to learn unified representations for broad audio understanding tasks, but it remains fragmented and is fundamentally bottlenecked by its reliance o…