Another day, another revolutionary AI tool is unleashed on the earth. Well, in this case, just announced during the Hot Pod Summit in Brooklyn this week. But regardless, Adobe has unveiled another major AI tool that should have not one, but several killer features that could reshape the landscape of music for video.
And, we guess, music for everything really.
Let’s take a look at Adobe’s upcoming Project Music GenAI Control and explore how this new prototype tool will be able to create music from text prompts as well as reference melodies. Plus, you know, explore how this could completely revolutionize the music industry for film, video, and beyond.
Adobe Project Music GenAI Control
Announced as a new prototype tool that aims to allow users to generate music using text prompts, and then edit said audio without needing to bounce out of your Adobe app and into some other dedicated editing software, this is pretty crazy stuff indeed.
From their announcement, Adobe has also shared that users will be able to input a full-text description (similar to other text-to-image or text-to-video prompts found in other AI models, we’d assume), and ask this AI to generate music in whatever style and with whatever parameters you’d like.
The tool will then offer other integrated editing controls to allow users to customize their AI creations adjusting elements like tempo and intensity, and even controlling things like patterns and structure. Sections of the songs will also be able to be remixed or generated as a loop so users will be able to better fit the music into their different projects.
Where Will the Music Come From?
Adobe has shared that this new Project Music GenAI tool is being developed in collaboration with the University of California and the School of Computer Science at Carnegie Mellon University. It sounds like the music used for the training is coming from the public domain.
This is important as Adobe will need to be very careful in where it sources its music especially as they plan to offer a reference melody feature where users will be able to upload a song, and then have the tool extend the audio clips or change elements to better fit your project.
However, from reports from The Verge, Adobe has not expanded upon where their audio is being sourced from and clarified how this would work exactly, especially if popular, licensed, and copyright-protected music is being referenced in this tool.
What’s Next for AI Music and Audio
This isn’t the only AI audio news of the week and it certainly appears that music and audio are in the crosshairs of AI technological improvements right now. Adobe has not yet shared when this new prototype tool might be released, or much less when it might make it into Beta for some of their apps like Premiere Pro or the like.
However, as with any timelines these days, things are moving fast. So it wouldn’t be surprising to see this tool make it into the hands of users sooner rather than later. And sooner rather than a competitor or other AI start-up cracks it first.
Author: Jourdan Aldredge
This article comes from No Film School and can be read on the original site.
