Adobe: Adobe has announced new updates to its AI-powered Firefly app, focusing on enhancing video editing tools. These updates include the ability to create sound effects based on user descriptions and voice recordings.
According to Qatar News Agency, the company noted that most current AI models lack audio support when creating clips, highlighting its efforts to bridge this gap with a feature that allows users to write a description of the desired sound effect and then record a voice simulation of it without imitating the voices. The company emphasized that the goal of this feature is to support creativity at the idea stage rather than achieving realistic simulations and to facilitate the timing of sound effects.
Other features revealed by Adobe include the Composition Reference feature, which allows users to upload an image or video to guide the video generation process. Additionally, the Video Presets feature allows users to select an output style such as anime, black and white, or line art. The Keyframe Cropping feature enables users to upload the first and last frame of a video and select its dimensions so that Firefly can produce a clip that adheres to that frame.
Last June, Adobe added support for external models, including the Veo 3 model, which Google unveiled at its I/O 2025 conference. Veo 3 is one of the few models capable of generating video with embedded audio. Adobe confirmed that all partner models are committed to not using user data to train their future models. Each image or video is digitally identified with the name of the model used to ensure intellectual property rights are protected.