fal· @fal · X·· 1 天前AI 评分64
AI 导读
fal 宣布 OmniHuman V2 正式上线。该模型可从单张图片和一段音频生成说话视频，同步口型与表情，并支持可选提示词来引导表演。
正文
OmniHuman V2 is now live on fal
- Generates talking videos from a single image and an audio track
- Synchronizes lip movements and expressions with speech
- Supports optional prompts to guide the performance
来源：fal · x.com