Alright, I've worked this problem before and this site is specifically designed to make it impossible to do what you're trying to do.

The next step, I fear, is to load the page with network diagnostics and analyze the traffic with the end of reverse engineering the API. After youve done that, maybe you can download the proprietary file type they use and then try to parse it.

There's a chance that someone already someone who's done this. I noticed a few fonts for music notation and a bunch of JavaScript.

From the design elements I see, it's loading some midi file or something like it, again, probably proprietary. I'm outta options.

Maybe try looking for a website that downloads directly from API or a GitHub that does the same.

Oof. Facepalm.

I see the issue now. I'll work on this and get back to you.