KlingAI releases Avatar 2.0, a digital human model. It can generate lip-sync videos up to 5 minutes long based on text content. The expressions are more realistic: laughter, anger, confusion, emphasis... all kinds of emotions can be expressed naturally, no longer as stiff as early AI characters. More agile movements: not just lip-syncing, but the overall performance is more like that of a real actor. Supports digital human video performances up to 5 minutes long.
Explanatory videos, promotional videos, music videos, storytelling, etc., are all supported. Various templates are available You can also upload your own image and voice to generate the generator.
