Hi, I make that site! We don't have any "basic effects like normalization/compression," so I think you may be mistaking us for something else?
Soundslice is oriented around sheet music that's synced with audio. So while we have a few audio-processing features (basic cropping, slowdown and fine-grained pitch correction), that stuff is all secondary to the notation aspect.
Thanks for sharing this. Your site is great. I've already learned a bunch of stuff, just browsing around the existing submissions.
I had a chuckle pondering whether you A/B tested "really freaking cool-looking" versus "really cool-looking" in the prompt. What a weird world we live in! :-)
Lol - I had a much fancier prompt to start, with things like "Be sure to invoke your frontend-designer skill" and "Make at least one applet inside the page with user-friendly controls".
But then I said screw it, let me try "really freaking cool"
Interpreting sheet music images is very complex, and I’m not surprised general-purpose LLMs totally fail at it. It’s orders of magnitude harder than text OCR, due to the two-dimensional-ness.
Obviously it will depend on what you're doing and one's tolerance for things many deem to be annoying. I've encountered people with incredibly high tolerance for slow and awkward workflows but, alas, I am not one of them.
If I had to call out one thing it would be that you can't do tests without having a database there. This results in incredibly slow tests for even the simplest things. I don't need to test database persistence every time I'm testing some domain logic. So maybe then don't do fat models and "map" the data from Django models to a domain layer? Well, congrats, you've just manually implemented a data mapper ORM, which is what SQLAlchemy is.
It works well for simple CRUD stuff, which is really useful. But it very quickly becomes a mess and a big ball of mud when you start to do more complicated things. IMO a db access layer or web framework should be completely independent of domain logic, but Django makes that really difficult.
> IMO a db access layer or web framework should be completely independent of domain logic, but Django makes that really difficult.
That is an issue/features in Django, depending on your view. You really don't get to do things the framework doesn't want. If you're trying to fight the ORM or any of the components in Django really, including the Django REST framework, you're going to lose and have a bad time.
There are certainly reason why you'd want to separate domain logic from the database access, but then Django isn't what you want. You're also going to miss out many of the things that makes Django easy to work with, like getting most CRUD operations for free.
I don't fully understand your comment, but Soundslice has had first-class support for tablature for more than 10 years now. There's an excellent built-in tab editor, plus importers for various formats. It's just the ASCII tab support that's new.
I’m not super familiar with Soundslice. But all the tab users I know use guitar pro or maybe ultimate guitar, and none of them can read standard notation on its own. Does Soundslice have a lot of tab-first users?
Yes, Soundslice has a ton of tab-first users. And in fact the primary reason I founded the site was to scratch my own itch of being able to create tab that's synced with real audio recordings. (I'm a guitarist myself.)
When I read the blog post, I thought it was saying that Soundslice didn't have any tab support.
The comparison of "we expect this (classical notation screenshot) but instead got this (ascii tab screenshot)" made me think that the only thing Soundslice supported was classical notation.
Definitely a subtle distinction there. Soundslice supports tab in many formats (MusicXML, Guitar Pro, PowerTab, TuxGuitar, PDF/images of published tab, or tab notated directly in the Soundslice editor) but didn't support ASCII format yet.
I wonder if LLMs will stimulate ASCII formats for more things, and whether we should design software in general to be more textual in order to work better with LLMs.
I've had AI create ascii-art (Nethack-style) dungeon diagrams when I asked it to write me a D&D adventure. Last time I tried it these dungeon diagrams were completely nonsensical, but that was a few years ago.
I think ASCII art in particular is generally a weak point for LLMs; maybe they could do better if they used an image model. Other ASCII syntaxes using matched { } delimiters seem like they would be easier.
Just a quick plug: check out Soundslice. It's interactive sheet music with a ton of learning tools built in, including easy navigation, looping, tempo changing and transposition.
We've also got a scanning feature that does OCR for sheet music, to get music into our system. Plus there's a full-featured notation editor. A good overview is at https://www.soundslice.com/features/
Soundslice (https://www.soundslice.com/) is a fully baked version of this idea. Tabs synced with videos, with a full notation/tab editor and tons of built-in music practice tools.
Soundslice is oriented around sheet music that's synced with audio. So while we have a few audio-processing features (basic cropping, slowdown and fine-grained pitch correction), that stuff is all secondary to the notation aspect.
reply