PhD student, The Chinese University of Hong Kong
1 paper at NeurIPS 2025
We propose a foundation model for unified speech generation with masked generative pre-training.
