Adobe said on the 21st that it officially released an audio feature for Firefly, its artificial intelligence (AI) image and video generation tool, that can create music, voice, and sound (effects).
The newly added features total three: music generation, voice generation, and sound effect generation. An Adobe official said, "With the release of this audio feature, we further strengthened Firefly's content generation AI capabilities."
The music generation feature, based on the Firefly music model, creates original music tailored to a video's length and mood. Voice generation, based on the Firefly speech model, turns text entered in natural language into natural-sounding speech. A key feature is precise control over voice, speaking speed, and emotion. Users can also access features from ElevenLabs, a voice AI startup.
For sound effect generation, it creates custom effects that match the actions and feel of the content.
Adobe designed audio generated with Firefly to be used commercially. From video to voice, users can now produce content in Adobe Firefly and apply it directly to finished works in one go.
Adobe also expanded the options for external AI models that integrate with Firefly. This time, it newly added Google's "Gemini Omni Flash." Now, in addition to Google, Kling AI, Luma AI, OpenAI, and Runway, users can also use Gemini Omni Flash.