Adobe Makes Firefly’s AI Audio Tools Generally Available

Adobe has made Firefly’s music, speech and sound-effects generators generally available, giving creators three separate ways to produce audio inside its creative AI suite. The release brings a soundtrack, a narrated script and timed effects closer to the image, video and design work Adobe already groups in Firefly.

The distinction is useful because the three jobs start from different inputs and require different controls. Generate Music makes original tracks around a video’s length and mood. Generate Speech turns a written script into a voiceover, while Generate Sound Effects produces custom sounds intended to follow the action, timing and energy of the content.

One studio, three audio roles

Each tool also has a distinct model arrangement. Music runs on Adobe’s Firefly Music Model. The speech tool uses the Firefly Speech Model and offers ElevenLabs as an option; Adobe says users can control voice, pacing and emotion. Sound effects use the Firefly Audio Model.

The production split

• Generate Music: original tracks designed around a video’s duration and mood.
• Generate Speech: script-to-voiceover generation with controls over voice, pacing and emotion.
• Generate Sound Effects: prompted sounds designed to match onscreen action, timing and energy.

Music carries the sharper business promise

That claim makes Generate Music more than a speed feature. Adobe is positioning it as a way to create a track for a finished project without separately sourcing music, while saying no separate subscription is required for tracks generated by the tool. The promise applies specifically to Adobe’s generated music, not as a blanket claim about AI audio products.

The result is a different proposition from a single general-purpose audio prompt box: Firefly separates music, narration and effects into tools built around their respective production tasks. Adobe’s broader pitch is that those tools sit alongside image, video and design capabilities in the same studio.

Two adjacent ways into Firefly

Adobe paired the audio release with two adjacent Firefly updates. It introduced a free Firefly AI Assistant experience with daily generations and added Gemini Omni Flash to the models available inside Firefly. Adobe says Gemini Omni Flash can take video, audio and image inputs alongside text.

Those additions expand access and model choice inside Firefly. The central product change, however, is the general availability of the three audio tools: a set of specialized generators whose practical appeal depends on whether creators value Firefly’s controls, integrated workspace and Adobe’s stated music-licensing terms.