Google Launches Veo 3.1 to Compete with OpenAI’s Sora 2

Google has introduced Veo 3.1, an updated AI video generator that enhances audio features and provides new editing capabilities for creators to better control their clips.

This release coincides with OpenAI's Sora 2 app gaining popularity in app stores and sparking discussions about AI-generated content on social media.

Google aims to position Veo 3.1 as a professional alternative to Sora 2's viral social feed approach. OpenAI's Sora 2, launched on September 30, reached 1 million downloads in five days and topped Apple's App Store.

Users can now create videos with synchronized ambient noise, dialogue, and Foley effects using the 'Ingredients to Video' tool, which combines multiple reference images into a single scene. The 'Frames to Video' feature generates transitions between images, while 'Extend' creates clips up to a minute long by continuing motion from the last second of an existing video.

Veo 3.1 shows significant improvement over its predecessor, handling coherence well across various styles. However, it prioritizes coherence over fluidity, making fast-paced action generation challenging.

In text-to-video mode, Veo 3.1 excels in generating coherent ambient sound that matches scene elements, with accurate dialogue and emotions. However, in image-to-video mode, it struggles with prompt adherence and reference images, prioritizing coherence too heavily.