Google introduced Veo 3, an AI model that for the first time allows creating synchronized video and sound from text in one step. Previously, AI systems could generate video without sound and audio from descriptions separately, but combining them required a second, manual stage. Trained by DeepMind, Veo 3 now generates dialogues, sound effects, background noise, and music together with the video, synchronizing the audio with the action and speech in the frame.
Alongside this, Google also presented Flow, a tool for generating 8-second clips using text and visual prompts. Flow builds a storyboard, “stitches” scenes together, and quickly turns ideas into short videos, freeing users from video editing routine. The updated Veo 2 allows changing camera angles and deleting objects during the generation process itself, while Imagen 4 generates images with photorealistic details and inserts readable text into images without errors.
These updates open new horizons for content creation by making it fast and accessible for everyone: they remove the technical barrier that previously prevented many creative ideas from becoming reality.
