Generative Extend overview

The Generative Extend tool, powered by Adobe Firefly, lets editors add those extra few frames to tell their story their way. Just click and drag to seamlessly add a few more frames to the beginning or end of a video or audio clip – to hold on to a character reaction for an extra beat, create a smoother transition, extend background sounds, or hide an unwanted camera movement. An internet connection is required to access Generative Extend as it utilizes a cloud AI model.

How it works

Current media requirements

Video 

These requirements are on the source media. Right now, you can’t extend media outside of these parameters, even if they are scaled in a sequence. 

• Video Source Media Resolution:
  • 1920x1080 or 1080x1920
  • 720x1280 or 1280x720
  • 3840x2160 or 2160x3840
  • 4096x2160 or 2160x4096
• 12-30fps
• 8 bit, SDR

Audio

• Generative Extend cannot create or extend spoken dialog. During extension, existing dialog will be muted to preserve the integrity of the original speech content.
• Due to the complexity of musical structures and potential copyright concerns, clips containing music are not eligible for extension.
• Currently, only mono and stereo audio are supported. Surround sound and 5.1 audio formats are not compatible with Generative Extend.

Managing Generated Content 

When you use Generative Extend, the AI-generated media is combined with your original clip to create a new clip in the Project panel. Each extension is saved as an H264 MP4 or .wav file in the Captured and Generated scratch disk assigned in your project settings. To set this location open Project Settings > Scratch Disks. Here, you can specify a new location for your generated media, allowing you to manage your storage according to your workflow and space requirements. 

Ongoing Development and Limitations

Generative Extend is the first application of generative AI for editors inside video editing software. It's an exciting tool to help many editors solve a common editing pain point, but this is just the beginning. We know that many workflows require higher-quality outputs and robust turnover processes. In the coming months, we'll expand the range of media types that can benefit from Generative Extend, including formats and color space. We're also working on reviewing the advanced workflow to make passing projects to other creatives seamless. 

Additionally, some limitations are known in this iteration with the following features:

• Generative Extend can only be applied to the beginning or the end of a clip, not both.
• Speech to Text transcription, Text-Based Editing and Captioning
• Media Intelligence
• Markers 
• Modify clip properties (Audio Channels, Color, Interpret Footage, Timecode) and speed adjustments
• Timecode and user-entered metadata 
• Interchange formats (XML, AAF, EDL, OMF) 
• Replace with After Effects Composition and Edit in Audition workflows 
• Multicam Source Sequences 
• Productions
• iPhone HDR clips

Learn more about these limitations and suggested workarounds.

Privacy and Data Usage 

Your media is never used to train Adobe's AI model. It's solely used to create the requested extension, ensuring your content remains private and under your control. Learn more about our generative AI commitments.

Future Improvements 

We're constantly working to enhance Generative Extend. Future updates aim to support higher resolutions, frame rates, HDR content, and additional aspect ratios, including vertical video for social media. 

Your Feedback Matters 

Help us improve! After each extension, right-click on the AI label and select Good Output or Poor Output. You can provide specific feedback on why the result worked well or needed improvement. 

• Generative Extend FAQs
• Generative Extend known issues
• User data storage for model output feedback

TALK TO US