Google has unveiled a new addition to its artificial intelligence (AI) toolkit, Vertex AI Media Studio, a series of generative tools that enable users to generate high-quality videos from a simple text prompt. The platform is founded on Google Cloud’s Vertex AI and merges several of the company’s cutting-edge models to manage all aspects of video production, including visuals, voiceover, and music. This is possible without any need for video editing or coding experience.
The process begins with the generation of an image using Imagen 3, Google’s image generation model. This image can be converted into a video using Veo 2, Google’s video generation model, which also provides customization tools. Veo enables users to choose the type of camera movement, adjust the frame rate, and determine the video’s duration. If the model introduces any unwanted elements, they can be removed using a Magic Eraser-style feature.
After the visuals are prepared, Media Studio employs Chirp, Google’s voice synthesis model, to add a voiceover. The production is completed with Lyria, a model co-developed by Google DeepMind and YouTube, which generates a music track for the background score. Ideally, the end product is a professional-looking video that can be shared