Video is one of the next frontiers in generative AI, and Adobe just teased a game-changing development. The creative software giant has given us a glimpse of a new AI video editing tool aimed at speeding up the workflow of its widely used software Premiere Pro.
This new tool will be familiar to anyone who uses Adobe's image editing program Photoshop. These effectively bring some of the key tools powered by Photoshop's Adobe Firefly to your videos, including Generative Fill, Generative Expand, and Delete.
New AI tools in Premiere aim to help video editors quickly and easily enhance shots and add or remove objects in a scene. These will be powered by new video models that join Adobe's Firefly Ai suite, which already includes images, vectors, designs, and text effects. The new tool headings are:
generative extend: Add frames to lengthen the clip using generative AI. Adobe says this breakthrough technology solves common problems that professional editors encounter every day, such as creating additional media to fine-tune an edit or adding shots for extra beats. or to better cover transitions.
Adding and removing objects: User can select and track objects and replace them. You can remove unnecessary items, change the actor's wardrobe, and instantly add set dressing, such as paintings on the walls or flowers on the desk.
text to video: Generate brand new footage directly within Premiere Pro by entering text prompts or uploading reference images. These clips can be used to ideate and create storyboards or create B-roll to augment live-action footage.
However, Adobe also previewed more features that may be added in the future. And this is a different and potentially more controversial approach to what Adobe has been doing with still images. The company is exploring options to bring third-party generated AI from OpenAI, Pika Labs, and Runway into Premiere Pro.
Daily design news, reviews, how-tos, and more, chosen by our editors.
Adobe has previously tried to distance itself from other generative AI tool developers by presenting Firefly as a more “ethical” and commercially safe AI model. That's because Firefly is trained using public domain assets and licensed images, rather than material collected from Adobe Stock. Web without consent.
Firefly caused a stir last week when it was confirmed that its training data included images on Adobe Stock generated by the AI model Midjourney. Now, when it comes to video, Adobe appears to be following the path that others like Canva have taken and allowing other AI models to be used within the software.
Adobe said in a press release: “Creative Cloud has always had a rich ecosystem of partners and plug-ins, and this evolution expands Premiere Pro as the most flexible and extensible professional video tool to fit any workflow. Adobe customers can create We want choice and endless possibilities when it comes to editing the next generation of entertainment and media.”
Adobe describes these as “early explorations,” and it should be made clear that there is no timeline for when such integrations will be generally available. However, the video preview shows how a video editor can use AI video generation models such as Premiere Pro's OpenAI Sora and Runway to generate his B-roll for editing into his projects in the future. . We'll also show you how to use Pika Labs in conjunction with his Generative Extend tool to add a few seconds to the end of your shots.
Adobe said, “While much of the early discussion around generative AI focused on competition between companies to produce the 'best' AI models, Adobe is looking to a more diverse future.” .
With a commitment to responsible innovation, the company is developing its own AI models and is “using what we learn to ensure that the integration of third-party models within our applications is consistent with our safety standards.” We plan to apply it,” he said. As one of the founders of the Content Authenticity Initiative, the company also promises to attach content credentials to assets created within the application.
Adobe also announced that Premiere Pro's AI-powered audio workflows will be generally available starting in May, including new fade handles, clip badges, dynamic waveforms, AI-based category tagging, and more.
Interactive fade handles allow editors to create custom audio transitions faster by dragging clip handles to create audio fades. Meanwhile, Essential Sound badges with audio category tagging automatically tag audio clips as dialogue, music, sound effects, or ambience and add new icons. Editors can instantly access the appropriate controls with one click.
Effect badges are new visual indicators that make it easy to see which clips have effects, quickly add new clips, and automatically open effect parameters from a sequence. Finally, the redesigned waveform intelligently resizes as the track height changes on the clip. The AI-powered Enhance Speech tool has been generally available since February to remove unwanted noise and improve poorly recorded conversations.
Find the lowest price for Adobe Premiere Pro in your area below, or check out our regular guide to Creative Cloud discounts.
