Sign In

Sick of AI-generated artwork? Then how about music?

Haebom
Adobe’s research team recently unveiled an experimental tool called Project Music GenAI Control, set to take the future of music creation to the next level. With this technology, users can generate personalized audio and music using text prompts, and then precisely edit them afterwards—holding the potential to fundamentally transform the music production process.
Nicholas Bryan, a research scientist at Adobe, mentioned that this project shows generative AI can become a true co-creator in the creative process. This opens up new possibilities for broadcasters, podcasters, and anyone else who needs custom music to make exactly what they want.
DITTO- Diffusion Inference-Time T -Optimization for Music Generation.pdf4.87MB
Adobe has also showcased safe and widely used AI in image generation with its generative AI model, Firefly. The company is working toward responsible AI development, and has even recently added a digital watermark that lets you check where generated content came from. All the image creation features you find in Photoshop and Express are powered by this model.
According to demo videos and recent research, with Project Music GenAI Control, users can create music from simple text prompts like “powerful rock,” “happy dance,” or “sad jazz.” The generated music can then be finely adjusted—tempo, structure, and intensity—along with editing options such as melody variations or making repeatable loops.
This tool moves away from traditional manual editing, letting users easily create and tweak just the part of the music they want. Bryan highlighted that it enables “pixel-level control” of music, allowing creators to shape and refine their audio with greater depth.
Just imagine if this were added to Adobe Premiere Pro or Audition—it would be incredible! Come to think of it, Adobe once had a service called Soundbooth... If this comes out, maybe making MIDI tracks could become even easier than with GarageBand.
If you get bored while waiting, try out Suno AI. It’s the most fun among T2A (Text-to-Audio) services—a clever tool that writes, composes, and guides vocals for a one-minute song complete with lyrics, all customized to the genre and mood you choose.
Subscribe to 'haebom'
📚 Welcome to Haebom's archives.
---
I post articles related to IT 💻, economy 💰, and humanities 🎭.
If you are curious about my thoughts, perspectives or interests, please subscribe.
haebom@kakao.com
Subscribe