1 paper
Dan Kondratyuk, Lijun Yu, Xiuye Gu +28
We present VideoPoet, a language model capable of synthesizing high-quality video, with matching audio, from a large variety of conditioning signals. VideoPoet employs a decoder-on…