The new video generation technology in YouTube shorts will help millions of people realize their creative vision
Artificial intelligence technologies (AI) to generate creative content quickly improves, but trouble -free ways of using them are still not widely available.
We change this and make these amazing technologies easier to access to millions of creators and billions of users around the world. Over the next few months, we transfer our advanced AI, VEO and Imagen 3 to YouTube creators Dream screen.
From our breakthrough Transformer architecture For years of research in diffusion models, these models are built almost a decade of innovation on Google and are now optimized for large -scale use.
During this trip, we worked closely with artists and creators and were guided by their curiosity and opinions so that our technologies would help more people in the implementation of their creative vision.
Creators can soon generate a video with a dream screen
By integrating VEO with the Dream screen, YouTube creators will soon be able to generate exciting backgrounds for their shorts. At the beginning of next year, we also allow you to generate 6 seconds of independent video clips.
Starting from the initial text line, Dream Screen uses Imagen 3 to generate four different images. The creators can choose a picture in the preferred style, composition or aesthetics from these options. Thanks to the chosen image, VEO generates a high quality 6-second background movie adapted to the creative vision of the user.
In 2025, the creators will also be able to use the Dream screen to generate 6 seconds of independent video clips for their shorts, such as underwater film revealing the Golden Gate bridge.
Increasing creativity for everyone
We hope that these new technologies for generating video and music will inspire more people to revive their ideas in a living, transformational way.
To help identify the content generated by AI, these creations will be marked with water using Synthid, and YouTube will apply label This clearly reports to viewers that he has been generated from AI.
We are still improving our generative AI models and expect help in building technologies that unlock creativity for everyone everywhere.
