Other Text-To-Video AI Projects

Several analysts are convinced that video is the next frontier in the generative AI field, and their convictions are largely backed by several text-to-video models that are actively being developed by the likes of Meta, Google, Amazon and other tech startups.

Sora is simply the latest to be introduced.

In 2022, Meta introduced a text-to-video AI generator called Make-A-Video. In a blog post announcing the project, Meta said that video generation tools could be invaluable “for creators and artists”, which is one of the reasons why they’re developing it.

Also in 2022, researchers at the Beijing Academy of Artificial Intelligence and Tsinghua University released a text-to-video model named CogVideo, which was the only publicly available text-to-video model at the time.

Still in 2022, Amazon introduced “Create with Alexa”, which is a feature that allows owners of Amazon Echo devices to come up with short animated stories with colorful graphics and background music.

Aimed for the pursual of children, it is basically an AI-backed story maker feature that allows kids to have a hand in the making of prompt-based animated short stories.

Last year, the Stability AI startup introduced the text-to-video-oriented Stable Video Diffusion project.

In a post introducing the project, the firm describes it as one that’s “powered by the latest advances in self-supervised learning”, and that it “capitalizes on massive neural networks to rapidly synthesize realistic videos for any prompt or reference material” which “enables endless new possibilities for visual storytelling and content creation across industries.”

In January this year, Google introduced Lumiere, which it describes as a “spacetime diffusion model for video generation”. The project can generate content on the text-to-video and image-to-video variables, as well as develop “stylized” content like cinemagraphs.

Sora was then introduced by OpenAI weeks after.