Adobe Announces Firefly Generative AI Video Model

A videographer must master both the process of shooting and the process of editing to create a compelling final video. During the shooting phase, you capture the scenes necessary to assemble the final project in the editing phase. Once you begin editing, your only option is to work with the footage you captured during shooting. If you neglected to film a particular shot, there is no way to create it later. Today marks a paradigm shift with Adobe's announcement of a Firefly Video Model in public beta, making it possible to create video footage from scratch.

Using Adobe's Firefly AI technology, video can be generated from a text prompt or an image. Users can change the camera angle and zoom in or out on the generated footage. Adobe assures creators that its AI models are safe from intellectual property lawsuits and that its AI model has been trained on Adobe stock footage, for which the company is the rights holder. The resolution, frames per second, and overall look of the AI-generated footage are designed with the professional video community in mind. Adobe expects widespread professional adoption of AI-generated clips.

AI-generated video is useful for creating additional B-roll footage, which can be used to cut away from the main narrative in a project. Adobe is taking this a step further by using AI technology to not only generate new clips but also extend existing ones. Adobe plans to integrate AI-powered video creation into all of its creative platforms, including Premiere Pro and the newly relaunched Firefly web application.

As illustrated in a promotional video supplied by Adobe, a user can use a text-to-image prompt in Firefly to generate an image, and the new Firefly Video Model can be used to animate that image into a video clip for use in Premiere Pro

There is also a new audio model launching in beta today that allows users to translate video into 20 languages. The AI technology modifies the original clip so the speaker’s lips are synced to the newly translated audio.

Adobe’s generative AI technology is accessed through the use of credits. As detailed in today’s press release, Adobe offers the following plans to enable full use of the new Firefly web app:

  • Firefly Standard plans are designed for creators looking to ideate. With early pricing on February 12, it offers 2,000 video/audio credits per month, up to twenty 5-second 1080p video generations, at $9.99 USD per month.
  • Firefly Pro plans are designed for creators looking to generate video/audio content more frequently. With early pricing, it offers 7,000 video/audio credits per month, up to seventy 5-second 1080p video generations, at $29.99 USD per month.
  • Firefly Premium plans will also be coming soon for professionals looking to generate video content on a regular basis, providing even more audio and video capacity for high-volume creators and teams.

The public beta version of the new generative AI video tool is available today on the Firefly web application.

John Ricard's picture

John Ricard is a NYC based portrait photographer. You can find more of Ricard’s work on his Instagram. accounts, www.instagram.com/JohnRicard and www.instagram.com/RicInAction

Log in or register to post comments