OmniTalker (Accepted by NeurIPS 2025)

This repo hosts the project page for OmniTalker.

OmniTalker: Real-Time Text-Driven Talking Head Generation with In-Context Audio-Visual Style Replication

Zhongjian Wang, Peng Zhang, Jinwei Qi, Guangyuan Wang Sheng Xu, Bang Zhang, Liefeng Bo

Tongyi Lab, Alibaba Group

<!-- [-ModelScope-blue.svg)](https://humanaigc.github.io/omnitalker/) -->

Citation

@misc{wang2025omnitalkerrealtimetextdriventalking,
      title={OmniTalker: Real-Time Text-Driven Talking Head Generation with In-Context Audio-Visual Style Replication}, 
      author={Zhongjian Wang and Peng Zhang and Jinwei Qi and Guangyuan Wang Sheng Xu and Bang Zhang and Liefeng Bo},
      year={2025},
      eprint={2504.02433},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2504.02433}, 
}