Wen Wu

SPEAR: A Unified SSL Framework for Learning Speech and Audio Representations

UniFlow-Audio: Unified Flow Matching for Audio Generation from Omni-Modalities

SciTS: Scientific Time Series Understanding and Generation with LLMs

Bayesian Speech Synthesisers Can Learn from Multiple Teachers

BrainOmni: A Brain Foundation Model for Unified EEG and MEG Signals

Enhancing Zero-shot Text-to-Speech Synthesis with Human Feedback

Bayesian WeakS-to-Strong from Text Classification to Generation