OpenAI Sora that can create one-minute videos from text will be available for everyone in a few months

OpenAI has confirmed that its new text-to-video generator platform – Sora – will soon be available to the public. The company’s Chief Technology Officer (CTO) Mira Murati revealed in an interview with The Wall Street Journal that Sora will be released “this year,” possibly within a “few months.”

Sora was unveiled
by OpenAI in February this year and it can
create lifelike scenes
based on text prompts. Initially, the tool was limited to visual artists, designers, and filmmakers. However, videos generated by Sora have already started appearing on platforms like X.

In addition to the public release, OpenAI plans to add audio capabilities to Sora eventually, aiming to enhance realism. Users will soon be able to edit the generated content to improve accuracy. Murati expressed OpenAI’s intention to make Sora a versatile tool for editing and creation. “We’re trying to figure out how to use this technology as a tool that people can edit and create with,” Murati said.

When asked about the data used to train Sora, Murati mentioned that it was from publicly available or licensed sources. While she didn’t provide specific details, she confirmed that Sora utilizes content from Shutterstock, with whom OpenAI has partnered. Murati also highlighted that Sora requires more resources, making it more costly to operate. However, OpenAI aims to keep the tool’s accessibility similar to DALL-E, their text-to-image model, in terms of cost.

“The model has a deep understanding of language, enabling it to accurately interpret prompts and generate compelling characters that express vibrant emotions. Sora can also create multiple shots within a single generated video that accurately persist characters and visual style,” says OpenAI.