@article{vividvoiceaunifiedframeworkforsceneaware,
  title = {VividVoice: A Unified Framework for Scene-Aware Visually-Driven Speech Synthesis},
  author = {Chengyuan Ma and Jiawei Jin and Ruijie Xiong and Chunxiang Jin and Canxiang Yan and Wenming Yang},
  year = {2026},
  eprint = {2602.02591},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2602.02591},
}