@misc{indiciaefb645e2048f4,
  title = {RASA: Replace Anyone, Say Anything -- A Training-Free Framework for Audio-Driven and Universal Portrait Video Editing},
  author = {Tianrui Pan and Lin Liu and Jie Liu and Xiaopeng Zhang and Jie Tang and Gangshan Wu and Qi Tian},
  year = {2025},
  url = {https://arxiv.org/abs/2503.11571},
  note = {Source identifier: 2503.11571}
}