Google Expands AI Video Creation with Veo 2 Integration

Users now have the exciting opportunity to generate standalone video clips for Shorts using artificial intelligence (AI) based on text prompts, starting today. This update is a result of the latest enhancement to Google’s Dream Screen.

Previously, users could only utilize text prompts to generate AI backgrounds for their Shorts, but this new functionality represents a significant upgrade.

The expanded features are made possible through the collaboration between Google DeepMind’s advanced video generation model, Veo2, and Dream Screen.

As highlighted in a YouTube blog post, "Need a specific scene but don’t have the right footage? Want to turn your imagination into reality and tell a unique story?" Users can now simply input a text prompt to create video clips that align seamlessly with their narrative. As Dina Berrada, YouTube’s director of product for generative AI creation, mentioned, “You can even create an entirely new world of content.”

Veo 2, in conjunction with Dream Screen, is capable of generating high-quality videos across various subjects and styles that correspond more accurately to users’ visions. Berrada also pointed out that Dream Screen has been optimized, now producing videos at a faster rate than before.

This facility of Veo2 addresses a common challenge known as "wallpaper footage". This term refers to situations where creators lack suitable video content that matches a particular storyline segment during video production.

While this AI-generated footage is currently limited to Shorts, it represents a notable addition to the growing suite of Generative AI tools available for public use.

This week, Adobe announced a new pricing model for its Firefly app, which allows users to create five-second videos for approximately A$0.82. They offer different subscription tiers, with the Standard plan at A$16.49 and the Pro version priced at A$49.49 per month.

Using text descriptions to create videos is becoming increasingly common, with platforms like Dream Screen and OpenAI’s Sora paving the way.

Additionally, third-party applications such as HeyGen enable users to have their scripts voiced by either a virtual version of themselves or various AI voice models. Users can choose from different voices, languages, accents, and even tones to customize their narration.

There are other valuable tools as well. For instance, Invideo AI allows users to submit a written story which it can transform into a full script, assign voices for narrators and characters, and find appropriate video overlays to fit the storyline.

Video editors can leverage Generative AI tools for specific needs within videos that have otherwise been produced without AI assistance. For example, they might select an AI voice for narration in a homemade documentary.

Google, AI, Video