During the "Made on YouTube" live event on Tuesday, YouTube announced a series of new generative AI tools for Shorts creators, aimed at simplifying and accelerating the short video creation process.
The core of this update is the integration of a customized version of Google's text-to-video generative AI model Veo3Fast. According to YouTube, this model can quickly generate videos at 480p resolution, allowing users to create video clips with simple text prompts, and it now supports recording audio.
Veo Features and AI Remix Tool
Veo's new features go beyond that. Creators can now apply actions from videos to static images, such as making people in photos dance to a dance from a video, achieving character animation. In addition, the tool supports adding different artistic styles (such as pop art or origami art) to videos, as well as adding characters or props based on text descriptions.
In terms of audio creation, YouTube introduced a new "Voice-to-Song Remix Tool", which uses Google's AI music model Lyria2, allowing creators to convert dialogue clips from videos into catchy background music. Creators can set different moods for the generated music, such as "relaxing," "danceable," or "funny," to create personalized soundtracks.
AI Editing Features Simplify Video Editing
To further simplify the editing process, YouTube also launched a new AI editing feature. This feature can automatically organize raw footage, select the best clips, and add music and transition effects to quickly generate a rough draft of the video. It can even automatically add voiceovers in English or Hindi for the video content, providing creators with an efficient starting point.
Currently, these new features are available in selected markets in the United States, the United Kingdom, Canada, Australia, and New Zealand. YouTube plans to gradually expand access to more creators and regions over the next few weeks and months.