Mark Zuckerberg's Meta has come up with a new AI model called Movie Gen, which can create realistic video and audio clips based on text inputs given by users.
With the new Generative AI model, Meta is challenging rivals OpenAI text-to-video model Sora and ElevenLabs's text-to-audio model.
What does Movie Gen do?
Unlike models created by OpenAI and ElevenLabs, Meta's Movie Gen integrates image, video and audio generation under one umbrella.
Movie Gen can create videos and audio clips using text inputs besides editing existing videos and even alter an image of yours into a unique video.
It can generate long high-definition videos based on text inputs at different aspect ratios. According to Meta, this is the first of its kind feature.
In order to alter an existing video, you can just give inputs like "dress the penguins in Victorian outfits". Besides you can create sound effects, background music and even soundtracks.