Adobe's generative artificial intelligence product Firefly has officially launched audio features that can create music, speech and sound effects.
Adobe said on the 21st that the music, speech and sound effect generation features, which had been offered as a beta version in Firefly, are now officially available.
Music generation uses the Firefly Music model. It creates original tracks tailored to the length and mood of the video a user is producing. A general license applies, allowing free use without concerns about takedowns for copyright violations.
The speech generation feature converts scripts into natural-sounding audio. Users can finely adjust the voice, speaking pace and emotion, and can choose not only Adobe's Firefly Speech model but also the ElevenLabs model. Sound effect generation creates customized sound effects that match the movements and timing of people and objects in a video. Audio generated with Firefly can be used commercially.
Adobe also added Gemini Omni Flash to its Firefly Partner Models. Firefly Partner Models is a feature that lets users select and use third-party generative AI models directly within the Firefly environment without leaving the Firefly app. As a result, users can now use Google's AI model in Firefly, in addition to Kling AI, Luma AI, OpenAI and Runway.
