Google has announced the launch of Vertex AI Media Studio, a suite of AI tools that lets users create videos based on text descriptions. The service is built on the Vertex AI platform and combines several advanced AI models to handle all aspects of video production, including visual effects, voiceover, and background music, without requiring users to have video editing or coding skills.
Image source: Steve Johnson/unsplash.com
Users are encouraged to start the process by creating an image using the AI generator Imagen 3. The resulting image can then be turned into a video using the Veo 2 algorithm, which also offers the ability to customize various parameters. According to Google, Veo allows you to choose the type of camera movement, such as drone or panorama, as well as adjust the frame rate and length of the video. If the algorithm adds any unnecessary elements to the video, they can be easily removed using the Magic Eraser tool.
Once the visuals are complete, the user is prompted to use the Chirp AI voice synthesizer to create a voiceover. Finally, the Lyria AI model, a joint creation of DeepMind and YouTube, will help generate the background music for the user’s video.
Theoretically, the end result should be a video ready for publication that is not inferior to a professional one, either in terms of what is happening in the frame or in terms of voice acting. And the user can create all this in one service, Vertex AI Studio, i.e. essentially the same service where developers test the latest versions of the Gemini AI model.