Google Brings Veo 3 Image-to-Video Generation Capabilities to Flow and Gemini AI Pro/Ultra

Google is making its Veo 3-based image-to-video generation capabilities available in Flow and Gemini AI Pro and Ultra, with experimental audio generation.

“Now, with a new photo-to-video capability in Gemini, you can now transform your favorite photos into dynamic eight-second video clips with sound,” Google’s David Sharon writes. “The explosion of creativity from users has been truly remarkable, with over 40 million Veo 3 videos generated across the Gemini app and Flow over the last seven weeks … your imagination is the limit when you create videos with Gemini.”

Google launched Veo 3 at I/O in May and then expanded its availability globally to all Gemini AI Pro subscribers last week. Since then, customers have made over 40 million videos with Veo 3 and the Flow app using text prompts. But Google is now rolling out the ability to generate videos–with an experimental audio generation feature–using images too. These videos include a watermark to show they are AI-generated and an invisible SynthID digital watermark, Google notes.

You can check out this capability at the Gemini website or using Flow. I don’t see this feature in Gemini yet, but I used the still image below to create the video at the top using Flow and its Frames to Video feature. Creepy and weird, for sure. (And for whatever it’s worth, the person in the front is British, so that generated voice is a bit off.)