Importing from PDFs
Soundslice cannot directly import PDFs. We rely on having knowledge of underlying music data, and a PDF doesn’t include that data.
A PDF is basically a photograph. It may look beautiful, but it doesn’t have enough information to let Soundslice do all the cool things Soundslice does.
But, there’s hope. If you have music only in PDF format, here’s what we recommend doing to get it into our system.
Vector PDFs vs. non-vector PDFs
The first thing to figure out is: is your PDF vector or non-vector?
Vector PDFs represent the graphics as instructions to the computer. Non-vector PDFs are essentially photographs.
If you generated your PDF from a notation program, it’s likely a vector PDF. If you generated your PDF by scanning or taking a photo of the notation, it’s a non-vector PDF.
If you’re not sure, here are two quick ways to determine whether you have a vector PDF:
- Using a PDF reader, zoom in continually. If the rendering remains crystal-clear and sharp, then it’s a vector PDF. Sometimes you’ll see individual lines/characters being rendered slightly before other characters — this is a telltale sign of a vector PDF. If things get pixelated/blurry as you zoom in, that means it’s not a vector PDF.
- Using a PDF reader, click and drag across the document. If you can select multiple bits of the page — for example, each notehead — then it’s a vector PDF. If it appears that only one thing is selectable — like it’s one huge image — then it’s likely not a vector PDF.
Have a vector PDF? That’s good news. It can be more reliably imported, because it doesn’t have problems that can trip up importers — like wavy staff lines, dust particles masquerading as augmentation dots and bad/inconsistent lighting.
We recommend the software PDFtoMusic Pro to import vector PDFs. Here’s how that works:
- Open your PDF in PDFtoMusic Pro, and let the software do its thing.
- In PDFtoMusic Pro, export as MusicXML.
- Import that MusicXML file into Soundslice.
(Our recommendation of PDFtoMusic Pro is genuine. We do not make any commission on this.)
Have a non-vector PDF? That basically means you’re dealing with a photograph of your notation. This is not ideal, but you can try using software that does so-called “optical music recognition” to extract the musical data from the image.
You may have heard of OCR, which is the concept of automatically extracting the words from scanned books. Optical music recognition (OMR) is the equivalent for sheet music.
There are several OCR products on the market, and none are perfect. This continues to be an unsolved computer-science challenge. But here are some products that attempt to do it:
(We do not make any commission on these links.)
With each of these products, you can export the scan as MusicXML. From there, you can import that MusicXML file into Soundslice.
If you use an automatic PDF import as discussed above, it will likely be imperfect. There might be some improperly detected notations throughout the file, depending on the nature of your music. Tuplets tend to confuse things, as do lyrics and chord names.
You may find that manually re-entering the music into a notation editor will end up being faster than cleaning up the importer’s mess. It really depends. Good luck!