Today, Google officially launched Veo 3.1, its latest video generation model. This version features significant upgrades in audio output, refined editing controls, and image-to-video conversion. As an iteration of Veo 3, released in May of this year, the new model further optimizes the fidelity of video generation and improves the accuracy of user prompts. This update not only maintains the core functionality of its predecessor but also significantly enhances the flexibility and expressiveness of video creation through breakthrough innovations in audio and object editing.
The key highlights of Veo 3.1 are the newly added audio support and object editing capabilities. Users can now add background music or sound effects to videos to enhance their content, while also allowing new objects to naturally blend into existing footage with a simple click, maintaining a consistent style. Google also announced that the video editing tool Flow will soon support the ability to remove existing objects from videos, further lowering the barrier to entry for professional-level video production. Veo 3 already supported advanced features such as character creation from reference images and generating complete videos from the first and last frames. This upgrade fully expands these capabilities with audio support, marking a new stage in video generation technology, moving from "vision-driven" to "audio-visual collaboration." Veo 3.1 is currently being deployed across Google's Flow platform, Gemini app, and Vertex AI service. Official data indicates that since Flow launched in May, users have created over 275 million videos, demonstrating strong market demand for AI-powered video tools.
