It’s been a year since OpenAI has announced Sora – its video and Audio deo generation model. OpenAI is now presenting the first major update with Sora 2, which can produce more real and physically accurate videos, with improved user control and editing capabilities.
https://www.youtube.com/watch?v=lecg6aj6dvy
Described as “GPT 3.5-contest”, Sora 2 also brings improved sound output with synchronized dialogue and sound effects. Sora 2 is capable of avoiding common problems associated with previous video generation models, such as morphing objects to distort the physical reality to complete the bugs and prompts.
Generated videos will also comply with multi-shot ts instructions and maintain consistency in various scenes. Sora 2 supports many video generation styles, including cinematic, real and anime.
OpenAI is also launching a single Sora iOS application with a single-only system. Sora 2 has several security, including restrictions on using public figures in Video Pay Generation. Open also confirmed that users can allow others to create “Camios” with their similarity, but they can be canceled by the user at any time in the future.
The initial rollout of Sora 2 begins today and it is the U.S. And will be limited to Canada. Speaking of prices, OpenAI confirmed that Sora 2 will initially be free with a “generous limit” so that users can explore their capabilities.
https://www.youtube.com/watch?v=1pawkvcjp0
ChatGPT Pro Subscribers High Quality Sora 2 Pro Model. Success. Open also confirmed his plans to release Sora 2’s API soon.
Original
