Google updates Veo 3.1 with full audio and editing features

Google has unveiled Veo 3.1, introducing audio and enhanced editing tools to its AI video creation suite. This upgrade enables users to adjust lighting, shadows, and tonal balance in generated clips, resulting in more expressive and coherent visuals. With “Ingredients to Video,” creators can supply up to three reference images, and the system will generate matching soundtracks along with visuals.

The “Frames to Video” feature enables the model to interpolate motion and audio between a starting and ending image, producing a seamless mini-clip. Using “Scene Extension,” users can lengthen an existing video by up to one minute, complete with new visuals and synchronized sound.

Google is also working on object removal: once live, creators will be able to erase unwanted elements and have the system reconstruct plausible backgrounds. These features are available now via a paid preview, accessible through the Gemini API, Gemini app, and Vertex AI platforms, with no change in price. Internally, Veo 3.1 is proving more consistent: it holds onto visual and audio continuity across edits and manages prompt adherence better than its predecessors.

Google positions this release as a stride toward closing the gap between AI-generated content and traditional video production systems. In short, creators gain refined control over both what is seen and heard in their AI videos, a move likely to raise the standard in automated content creation.