METHOD AND SYSTEM FOR CREATING FACIAL EXPRESSION BASED ON TEXT Russian patent published in 2020 - IPC G06T13/40 G10L21/10 

FIELD: data processing.

SUBSTANCE: invention relates to processing image data. Data of at least one speech signal is obtained. Voice information sections are divided into time windows. A frequency spectrum image is formed for each time window to obtain a sequence of frequency spectrum images. Based on the sequence of images of the frequency spectrum, a sequence of data on the plurality of coordinates of the points forming the face mask is determined. Face mask is placed on 3D-model of head to form sequence of frames containing image of 3D-model of head with face mask arranged on it. A series of frames of the dynamic texture of the face mask is generated based on a sequence of images of the frequency spectrum. A sequence of frames containing an image of resulting 3D model of the head with a dynamic texture of the face mask is formed.

EFFECT: technical result consists in enabling generation of a video stream with an animated image of 3D head model with a dynamic texture of the face mask on the basis of the speech signal data.

9 cl, 4 dwg