Google’s AI video generator is getting better editing and more audio
Key Takeaways
- Google updated its AI filmmaking tool, Flow, with the Veo 3.1 update, adding advanced lighting and shadow editing.
- New features enable audio generation in videos, including creating clips from reference images ('Ingredients to Video').
- The 'Frames to Video' feature creates smooth transitions between start and end images with accompanying audio.
- The 'Scene Extension' feature allows users to add up to a minute of generated video and audio to existing clips.
- Google plans to soon allow users to remove any object from a video, with the AI restructuring the scene accordingly.
Google is rolling out significant upgrades to its AI video creation tool, Flow, centered around the Veo 3.1 update, which enhances realism by allowing users to precisely control shadows and lighting within AI-generated videos. A key aspect of the update is the expansion into audio generation, enabling features like creating videos from three reference images, dubbed “Ingredients to Video,” and smoothly transitioning between start and end images with audio via “Frames to Video.” Furthermore, the “Scene Extension” feature allows users to seamlessly append up to a minute of new, audio-accompanied footage to the end of a clip. Google also previewed a future capability allowing users to remove any object from a video, with the AI intelligently reconstructing the background. Veo 3.1 is available to developers via the Gemini API as a paid preview and is enabled within the Gemini app.
