YouTube Announces Generative AI Tools for Shorts - New Features

YouTube Introduces Advanced AI Tools for Shorts Creation

During its Made on YouTube event held on Tuesday, the platform announced the release of innovative generative AI tools designed specifically for Shorts creators. These new offerings include a customized version of Google’s Veo 3 text-to-video AI model, a novel remixing capability, and an “Edit with AI” function, among other enhancements.

Veo 3 Fast: Accelerated Video Generation

The tailored iteration of Veo 3, designated Veo 3 Fast, is engineered to produce outputs with reduced latency at a resolution of 480p. This facilitates the streamlined creation of short-form video content, according to YouTube. Notably, sound can now be incorporated into generated videos for the first time.

Initial availability of this update is limited to users in the United States, the United Kingdom, Canada, Australia, and New Zealand. Future plans involve extending functionality to additional regions globally.

YouTube is also expanding the capabilities of Veo within Shorts. This includes the ability to transfer motion from video footage onto static images.

Animating Images and Applying Styles

For instance, a still photograph can be animated by replicating the movements of a person performing a dance captured in a video. This functionality is enabled by technology that accurately captures and transfers movement between subjects.

Creators can now leverage Veo to apply diverse artistic styles to their videos, such as pop art or origami aesthetics. Furthermore, the ability to introduce objects – including characters and props – via text descriptions has been added.

These enhanced features are scheduled for rollout over the coming months.

New Remixing Tool: Speech to Song

A new remixing tool empowers creators to transform dialogue from eligible videos into engaging soundtracks for other Shorts. This allows for the repurposing of memorable lines and sounds.

“YouTube serves as the world’s premier creative environment, a hub for emerging trends and inspiration,” stated Dina Berrada, Director of Product, Shorts and Generative AI Creation, in a blog post. “Imagine encountering a dialogue snippet that ignites an idea – a humorous phrase, a striking quote, or a unique sound – and wanting to integrate it into a new composition. Our Speech to Song remixing tool enables precisely that.”

The feature utilizes Google’s Lyria 2 AI music model to generate the soundtrack. Creators can then customize the song’s mood with options like “chill,” “danceable,” or “fun.”

Testing of this feature will commence shortly, with a broader rollout to creators in the United States planned for the coming weeks.

Edit with AI: From Footage to First Draft

The new “Edit with AI” feature assists creators in transforming raw footage into initial drafts. It intelligently identifies and arranges the most compelling moments, adds music, and incorporates transitions.

The tool can even generate a voice-over that dynamically responds to the video’s content, available in both English and Hindi. YouTube intends this feature to provide creators with a convenient starting point for their Shorts projects.

Currently, YouTube is conducting experiments with “Edit with AI” on both Shorts and within the YouTube Create app. Expansion to select markets is anticipated in the weeks ahead.