识别文本内容以及语音内容让人物做出一些相符的动作。比如做某种产品的介绍时，人物的手部的动作会指向产品或者是指向某一方向。给后期的人员添加产品，更加的自然。

(Identify text content and speech content to enable characters to perform corresponding actions. For example, when introducing a product, the character's hand movements should point to the product or a specific direction. This will make it more natural for post-production staff to add the product.)