Yuzhu Wang

A training-free dynamic clustering method significantly improves long speech separation performance, especially in sparse scenarios with unknown speaker counts.

Executable Blender code transforms text-to-video generation, enabling unprecedented control over scene dynamics and visual fidelity.

Achieving a breakthrough in music source restoration, DTT-BSR+ significantly enhances signal quality while preserving semantic integrity, outperforming existing methods.

A hybrid GAN architecture with RoPE transformers and dual-path RNNs achieves state-of-the-art music source restoration with only 7.1M parameters.