Adobe's new AI tool can generate audio using text prompts
Adobe has unveiled Project Music GenAI Control, a cutting-edge AI tool that enables users to create and personalize music, sans any professional audio expertise. By simply typing in text prompts like "happy dance" or "sad jazz," users can generate music in their desired style. The tool also offers built-in editing controls for customizing the audio, such as altering repeating patterns, tempo, intensity, and structure.
How Project Music GenAI Control works
Unlike Google's MusicLM and Meta's AudioCraft, which only permit users to generate audio with limited editing options, Project Music GenAI Control provides more advanced editing features. This means users won't have to keep generating audio until they get the desired result or use separate audio editing software. Adobe's Senior Research Scientist, Nicholas Bryan, described the new tool as offering "a kind of pixel-level control for music."
Comparison with existing tools
With Project Music GenAI Control, users can also remix sections of music and create repeating loops for backing tracks or background tunes. Adobe claims that the generated audio can be adjusted "based on a reference melody" and extended for fixed animations or podcast segments. This innovative tool aims to bridge the gap in the market by providing a more comprehensive music creation and customization experience, similar to Photoshop for music.
What about its availability?
Adobe has developed Project Music GenAI Control in collaboration with the School of Computer Science at Carnegie Mellon University and the University of California. Adobe has described the tool as an "early-stage" experiment. Hence, its features may eventually be introduced in the firm's existing editing tools like Audition and Premiere Pro, but it will take time. We do not know when it will be introduced to the public.
