Social media giant Meta has unveiled ‘Movie Gen’, a new AI model that is capable of turning user prompts into realistic-looking video and audio clips.
Meta’s Movie Gen follows OpenAI’s launch of Sora, a ground-breaking tech innovation that enables users to create videos from text prompts. With this new tool, Meta aims to transform content creation.
Few samples of Movie Gen’s creations feature stunning videos of animals swimming and surfing, as well as personalized videos of people engaging in activities such as painting on a canvas with real face.
In addition to generating high-quality videos, Movie Gen can also be used to edit existing videos and still images. Furthermore, background music and different sound effects synced to the videos can be generated using text-prompts, according to Meta’s blog post.
The tool can generate videos in various aspect ratios. The exiting videos style, transitions and other elements can be altered using text inputs. The AI model enables users to create videos up to 16 seconds and audios up to 45 seconds.
Meta also highlighted Movie Gen’s exceptional performance that it functions better than the AI tools developed by OpenAI, ElevenLabs, and Runway. Though it’s exciting and promising, the tech giant won’t be rolling out this innovative tool in near future.
Despite its vast entertainment potential, Meta spokespeople revealed that the company will restrict Movie Gen’s access to public, unlike its Llama series of large-language models.
This new announcement comes amid the pressing debates revolving around the use of AI, specifically about copyright infringement. Recently, the leading actor Scarlett Johansson accused the ChatGPT for using her voice without consent for its chatbot.
On the other hand, Lions Gate Entertainment, the company known for its popular film series titled “The Hunger Games” and “Twilight,” revealed that it is giving AI startup Runway access to its film and television library to train an AI model in exchange for tools to use in film production.
