Adobe Firefly can now generate sound effects from your audio cues

Adobe has introduced new features to its Firefly generative AI hub, focused on improving video capabilities. The key updates include: 1. Sound effect generation: Users can now describe a sound effect and record themselves making the sound, allowing Firefly's model to reproduce the effect at the desired intensity and timing. 2. Composition Reference: Users can upload a video or image to guide the generation process, and use Video Presets to define the style of the final output, such as anime, black and white, or vector art. 3. Keyframe Cropping: Users can upload the first and final frames of a video and select an aspect ratio, and Firefly will generate a video that stays within the desired format. Adobe has also added support for more third-party models, including Veo 3, one of the few AI models that can generate video with sound. The company continues to prioritize data privacy, ensuring that Firefly customers don't accidentally use copyrighted material.