Adobe adds new AI audio generation tools for music, speech, and sound effects to Firefly

Adobe has announced that Firefly, its suite of AI-powered creative tools, can now generate music, speech, and sound effects, expanding the platform beyond images and video. Generate Music uses the new Firefly Music Model to create original tracks based on a project’s mood and duration, with music that can automatically match a video’s length. Adobe says the generated tracks are universally licensed and safe for commercial use, positioning Firefly as an alternative to other AI music generators like Suno and Udio, with a stronger focus on commercial content and licensing.

Generate Speech can also turns written scripts into AI voiceovers with controls for voice, pacing, and emotion, using either Adobe’s Firefly Speech Model or ElevenLabs models. Generate Sound Effects uses the Firefly Audio Model to create custom effects that match the timing and energy of a piece of content, covering use cases such as social clips, vlogs, tutorials, short films, and podcasts.

Its integrated Firefly AI Assistant, still in beta, is also getting expanded natural language tools for tasks such as creating brand kits, storyboards, and batch image edits, along with a new free experience offering daily generations. Adobe is also adding Google’s Gemini Omni Flash, which accepts text, image, audio, and video inputs and can turn rough concepts into storyboarded video cuts.