Adobe Firefly continues to evolve rapidly, with a fresh wave of updates aimed at enhancing its video capabilities. Since its major redesign in April, Adobe has been aggressively rolling out new features.

The latest update brings some powerful tools for content creators, including sound effect generation and deeper video customization options.

These improvements reinforce Adobe’s commitment to growing Firefly as a competitive hub for generative AI tools in visual media.

🔗 Adobe Firefly can now generate sound effects from your audio cues – Engadget
🔗 Adobe’s new AI agent can show you how to use Photoshop – Engadget
🔗 Adobe rolls out generative AI video tools in beta – Engadget

Sound Effects with a Personal Touch

One of the standout features in this update is the ability to generate sound effects based on user prompts and recorded cues.

Users can describe a sound—such as a zipper—and then make the sound themselves, which helps Adobe’s model understand the desired timing and volume.

In a demonstration, an Adobe employee described and performed a zipper sound. The tool then reproduced it faithfully with correct timing and intensity. A timeline editor at the bottom of the interface enables precise placement of audio to video.

However, the feature still has room to grow: “The translation was less convincing when the employee used the tool to add the sound of footsteps on concrete,” Engadget reported.

Uploads, Cropping, and Stylization: A Creative Toolkit

The update also brings greater control over how videos are generated:

Other New Additions:

• Composition Reference – Upload your photo or video to guide the generation process
• Video Presets – Choose from styles like anime, black and white, or vector art
• Keyframe Cropping – Upload a first and final frame, select an aspect ratio, and Firefly generates video within that format

These additions give users more stylistic and format-based customization, improving creative control without needing professional video editing expertise.

https://usaherald.com/trump-files-20-billion-libel-lawsuit-against-wall-street-journal-over-epstein-allegations/

Partnerships and Privacy

Adobe also announced support for Google’s Veo 3, one of the few AI models capable of generating video with synchronized sound. As with other partner models, Adobe confirms that data generated within Firefly won’t be used for training future models, protecting user privacy.

Vice President of Product Management Zeke Koch said, “We're relentlessly shipping stuff almost as quickly as we can,” adding that Adobe will only integrate third-party models that respect Adobe’s data terms.

Each asset created in Firefly is digitally signed with embedded data about the AI model used. These “content credentials” can be verified through Adobe’s Content Authenticity web app, helping creators ensure safe commercial use.