Sora is a text-to-video tool released by OpenAI on February 15, 2024. It is an AI video model designed specifically for generating videos.
Essentially, Sora is an AI model, but unlike others, it is tailored for video generation.
The technology of generating short videos based on descriptive text input is not new. In our country, there are apps like Jianying Pro that offer similar functionality. You write a piece of text, and then the app automatically generates a video based on your text.
The concept of text-to-video has been around for a while, so why has Sora generated such a huge response?
I believe the key lies in the quality and effects of the generated videos.
If you’ve used Jianying Pro’s text-to-video feature, you might have noticed that it mostly consists of overlaid images, occasionally supplemented by snippets of relevant video footage sourced from the internet. However, the quality and effect of the generated videos are often mediocre.
Sometimes, it’s evident that there isn’t a strong correlation between the text and the generated video; it feels like a simple collage.
This lackluster performance is characteristic of most text-to-video AIs.
The reason Sora has garnered widespread attention is because of the quality of the videos it generates.