Meta Unveils New AI Model That Can Generate Instant Videos With Added Sounds

Meta, the owner of Facebook and Instagram, is here with a new AI image generator. The tech giant announced on Friday that it had built a new Artificial Intelligence model called Meta Movie Gen. Meta is among the many companies that are building this technology.

In the yearly quarter of 2024, the artificial intelligence company OpenAI unveiled a technology called Sora. Sora lets people generate photorealistic videos, like woolly mammoths trotting through a snowy meadow — simply by typing a sentence into a box on a computer screen.

However, due to some initial concerns that the technology might be misused to generate disinformation or for malicious activities, it wasn’t released to the public. The AI model was available to only a small group of testers. Since then, many companies have been planning to release similar technology. 

Meta is here with Meta Movie Gen, which is capable of creating realistic-seeming video and audio clips in response to user prompts. The company claims it can rival tools from leading media startups like OpenAI and ElevenLabs. 

Just by giving a brief description, one tool can create a video. And giving another command for sound, the second tool can create audio.

Samples from Movie Gen showcased people creating realistic videos of animals like koalas surfacing in the ocean, a snake slithering through a forest, and a person riding an all-terrain vehicle across the desert. It also generated clips using people’s authentic photographs to depict them performing actions like painting on a canvas. Each of the videos created had music as well as sound effects.

“Video is nothing without audio,” Ahmad Al-Dahle, Meta’s vice president of generative A.I., said in an interview.

According to some speculation, there is a bittersweet feeling among the people. Many thought this could ease the works of Hollywood filmmakers but could potentially harm the less experienced digital artists. Others worry that this will be a target to generate disinformation, making it difficult to comprehend what’s real on the internet.