Google Photos develops AI video editing feature Soba

en.Wedoany.com Reported - Google Photos is developing an AI-powered video editing feature, codenamed "Soba," which leverages the Gemini Omni multimodal model to allow users to perform conversational editing and transformation of videos through text or voice commands.

App reverse engineering expert AssembleDebug discovered the Soba button in the "Create" tab of the Google Photos Android app, with an icon resembling the YouTube video icon and featuring Gemini's signature sparkle pattern. Internal test strings indicate that when Soba is enabled, the existing "Remix" button is automatically renamed to "Photo Remix" to avoid confusion with the new "Video Remix" feature.

Gemini Omni is Google's latest video generation model, supporting multimodal context and conversational editing. For example, users can instruct it to convert personal videos into a claymation style, and the model can automatically identify specific events in the footage and apply effects accordingly. Currently, Google Photos does not include the complete code required to run Soba, but the button trigger logic has been pre-configured.

This feature differs from the existing "Photo to Video" tool, which only converts static images into videos. Soba will take existing videos as input and output modified videos. When the Photo Remix feature launched in July 2025, it initially offered only four presets, expanding to 13 five months later. Therefore, analysts expect Google to adopt a cautious approach with Video Remix as well, potentially offering only limited presets initially rather than full conversational editing capabilities. Photo Remix uses the Nano Banana model, while Soba is likely to implement similar functionality based on Gemini Omni.

This article is compiled by Wedoany. All AI citations must indicate the source as "Wedoany". If there is any infringement or other issues, please notify us promptly, and we will modify or delete it accordingly. Email: news@wedoany.com