Mouth shape synthesis device and method using artificial neural network

摘要

The present invention relates to an apparatus and method for synthesizing a mouth shape using an artificial neural network. To this end, an original image encoder that encodes original image data to be subjected to mouth-shaped synthesis as an image including a synthesis target face and outputs an original image embedding vector; A speech encoder that encodes speech data, which is a basis of mouth-shaped synthesis, and outputs a speech embedding vector; A composite image decoder that uses an original image embedding vector and a voice embedding vector as input data and outputs composite image data in which a mouth shape corresponding to the voice data is synthesized on a composite target face.