[DigitalToday reporter Chi-gyu Hwang (황치규)] Google has added a feature to its video editing app Google Vids that lets users direct avatar actions and change their appearance using text prompts, TechCrunch reported on Wednesday.

The update also adds support for Veo 3.1, direct export to YouTube and a Chrome extension-based screen recording function to Google Vids.

Users can use natural-language prompts to instruct an avatar to “act” in a scene. They can also set up situations where the avatar interacts with products, props and equipment. Google said it keeps character consistency even when the output changes dynamically.

Google also supports refining a character’s appearance, changing outfits and creating new backgrounds with prompts to match a video’s theme.

Vids generates 8-second clips with Veo 3.1 inside its editing tools. All users get 10 free generations per month. Google AI Ultra and Workspace AI Ultra accounts can generate up to 1,000 Veo videos per month. Google added the Lyria 3 and Lyria 3 Promusic models to Vids last month, also enabling users to add sound effects and music to clips.

Finished videos can be exported directly from Vids to YouTube. Exported videos are private for now. Google also added a Chrome extension for screen recording to the Vids product lineup, allowing users to capture the screen with audio or video.

Google has expanded Vids features since its release in 2024, targeting demand for corporate content production. It introduced AI avatars last year. In February this year, it added 2D and 3D cartoon-style avatars and support for seven voice dubbing languages: French, German, Italian, Korean, Portuguese, Spanish and Japanese.