Generative AI & Music Notation - It’s Here.

By now you are probably familiar with Generative AI apps, specifically for music, and how they are slowly being introduced to the music classroom. Tools like PracticeFirst, Sight Reading Factory, and Moises.ai are already reshaping how we approach practice, assessment, and ensemble preparation. For the past two years, music educators who have attended my talks on Generative AI have asked me about tools that might generate music notation, and until very recently, I haven’t been able to point them to anything. Recently however, a new tool called NotaGen came up in my LinkedIn feed that promised to do just that. Created by a group of While it is still in its infancy, it is really impressive. NotaGen is a web-based generative AI tool designed to create musical notation automatically based on user input and parameters. Think of it as ChatGPT meets Flat or Sibelius. With just a few clicks and prompts, NotaGen can generate original sheet music, complete with customizable instrumentation, time signatures, key signatures, and more. Even better, it outputs music notation directly in SVG format, making it easy to embed and share. Pretty cool. Here is a sample composition for orchestra created by this new tool:

So how does it work? At the moment, NotaGen lives on a development platform called GitHub so it really looks like you’re coding. There are three prompt windows: Period, Composer, and Instrumentation. I decided to experiment with this new tool by selecting Classical, Beethoven and Chamber Ensemble. What happened first is that a series of fast moving text-based code appeared in the Generation Process window. It looks like this: [r:120/7][V:1](B/d/f/=e/ f/d/b/a/ b/f/_e/d/)|[V:2](B,/D/F/D/) (B,/D/F/D/) (B,/D/F/D/)|[V:3]B,2) z2 z2|[V:4]B,,2) z2 z2|. It looks like it’s giving you the tempo (120) and then the instrument (violin) and then a series of note names. After about 5 minutes, the generation was done and then the score was rendered in the Post-Processed ABC Notation Scores window (it uses a notation engine called ABC notation). Finally, an image of the score appeared as well as an audio file. This is what I ended up with:

Here is a link to the full score in PDF format. And this is what it sounds like:

Is it amazing music? Not really. Does it sound like a chamber work for string quartet by Beethoven? Sure. Am I impressed by the output - purely from a technological standpoint? Absolutely.

So why should music educators pay attention? As music teachers, we’re always on the lookout for tools that inspire creativity, save time, and make music accessible to students of all levels. NotaGen could be a big step forward on a few different fronts. Imagine you’re teaching middle school general music, and you want students to engage in a composition project. Some kids are naturals with notation software; others struggle to get ideas onto the page. NotaGen can act as a creative spark—students can generate a melody or chord progression, then tweak and build on it. It lowers the barrier to entry and helps every student feel like they can be a composer. Is it really composing? I would argue that prompting and composing are two very different things, but this could serve as inspiration for students own original works - exactly the same as tools like Udio and Suno can be used.

In both middle and high school music theory classes, Notagen can serve as an example generator for analysis. Need fresh examples of cadences? Want students to analyze different harmonic progressions? Because the three eras of classical music that the AI model has been trained on stick pretty closely to a set of compositional “rules”, the works that are created are really similar to the 1.5MM works that they were trained on in my opinion. Using this tool you can quickly create endless examples for analysis or part-writing exercises. It’s like having a teaching assistant that never gets tired of writing new material.

High school band and orchestra directors are always hunting for fresh sight-reading material for their ensembles. I think that the works that NotaGen can generate might serve as a generator for simple or complex exercises on the fly. While nothing like the incredible sight reading tool Sight Reading Factory (which is much more versatile and customizable than NotaGen), you can have Notagen generate examples instantly—and then export them for a quick read-thru - and who knows - it might even be good enough to perform.?

The implications for professional and amateur composers are equally exciting—and, to be honest, a little scary. On the one hand, NotaGen offers composers a tool for generating raw material quickly. Stuck on an idea? Let NotaGen generate a few melodic sketches to get the creative juices flowing. It can serve as a starting point for larger works, freeing composers from the tyranny of the blank page. On the other hand, this type of technology raises important questions about authorship and originality. If an AI generates a melody that inspires a piece, who “owns” it? For now, NotaGen is open-source and explicitly a demo, but as these tools become more sophisticated, these questions will need to be addressed by the music community at large.

What excites me most is the potential for composers to collaborate with AI in much the same way visual artists are doing today. Rather than viewing AI as a competitor, we can embrace it as a creative partner—one that speeds up certain processes, offers novel ideas, and helps composers focus on the parts of music-making that require a human touch: emotion, interpretation, and nuance.

Right now, NotaGen is a simple browser-based tool. You visit the site, set your parameters, and let it generate music. It’s not really ready for a large audience and it doesn’t yet integrate playback or MIDI export. There are some other similar apps, like Klangio that can actually transcribe music from audio (more to come on that). My personal interaction with the software was pretty clunky but for an early-stage demo, it’s impressive—and for now it’s free! Generative AI in music is still in its early days, but tools like NotaGen show us where things are heading. For music teachers, this could mean faster lesson prep, more creative student projects, and deeper engagement with the compositional process. For composers, it’s a powerful new tool in the creative toolbox. As always, the best way to understand its potential is to dive in and try it yourself. Explore, experiment, and see how NotaGen might inspire your next class—or your next masterpiece.