|
|
|
|
|
by adrianh
823 days ago
|
|
At Soundslice we're using a custom machine learning system to convert sheet music images into semantic music data: https://www.soundslice.com/sheet-music-scanner/ Personally I think the user experience is interesting because we show you very specific questions for low-confidence decisions. Some example screenshots are on that link above. Over time, the number of manual questions has gone down, as our models have gotten smarter about the (seemingly endless!) edge cases in music notation. Once music is uploaded and scanned, you can use our bespoke notation editor to make any edits. The original image is tightly integrated into our editor, so you can cross-reference. This is my first production ML product after 20 years of being a web developer. I wrote up some general thoughts here: http://www.holovaty.com/writing/machine-learning-thoughts/ |
|