Google Unveils Veo 3.1 AI Video Generator with Enhanced Image-to-Video and Synchronized Audio Generation

Google has announced Veo 3.1, the latest version of its AI video generation model, highlighting improved prompt adherence and image-to-video conversion when paired with textual descriptions. Veo 3.1 is now available for preview through the Google Gemini API and has been fully integrated into Google’s Flow video editing suite.

Built upon Veo 3, introduced earlier this year at Google I/O, the updated model delivers enhanced performance in following user prompts and can more seamlessly transform uploaded image assets into videos. A new feature enables simultaneous image-to-video conversion with audio generation, a capability absent in its predecessor, and within the Flow editor it powers an “Image-to-Video” option that lets users set a starting and ending frame to guide intermediate footage.

According to Google’s shared samples, videos produced by Veo 3.1 still retain a faintly artificial quality, with results varying by prompts and subject matter, and it is claimed not to rival OpenAI’s Sora 2. Flow distinguishes itself from Adobe Firefly by its ability to generate synchronised audio, which also extends to the editor’s video extension and object insertion features.