New to Musescore - Scanning / conversion issues
I am new to Musescore. Trying to do a basic Scan, PDF Upload, Download, transpose.
The document received has little resemblance to the image scanned and uploaded.
Using a good quality Epson scanner, scanning in high resolution B/W.
The document attached is a sample of the problem.
1st two pages are the upload PDF. 2nd two pages are what I got back from Musescore.
Guidance appreciated.
JB in Chicago
| Attachment | Size |
|---|---|
| dennis2.pdf | 520.84 KB |
Comments
You are asking too much from a damaged photocopy of a photocopy.
I am a great fan of optical music recognition (OMR). Only today I captured a 2-page score for SATB choir and organ using OMR (commercial, not freeware). The recognition was almost 100% perfect: notes. lyrics, dynamics - everything. But I did provide a PDF which was a pin-sharp image.
Sot with your PDF I think that it would be much quicker just to typeset the entire score by hand in MuseScore.
I've tested a few music recognition programs. None of them are perfect in any way. However, if your goal is to have the music in Musescore, my experience it's better to correct the recognition before opening in Musescore.
In the example her, I used Audiveris, that's the same software behind the Musescore "upload PDF" but installed as a stand alone SW on my computer. Here you have a little better control of the recognition process an you can make corrections before it' exported to XML. 1. - 3 are snapshots from Audiveris
In the files below, you can get the progress.
1. JustAfterRecognition.png
2. SomeCorrectionsLater.png. By the way, the red shaded measures are timing errors which you will see as s grey - or + sign in the upper right corner above the corresponding measure in MuseScore.
3. AllRecognitionErrorsRemoved.png
4. Audiveris export to XML PDFs.mxl
5. View from MS Musescore4x.png
6. in MS format PDFs.mscz
In fact, you won't find all errors immediately but you can make regular export to XML during the process and check the result in Musescore.
The PDFs were not of very high quality, hence initially quite a number of errors. In some cases, I think it would be better to crop and remove items that are not recognised anyway, i.e. title and lyrics.
All in all, I spend about 10 minutes on one page but on the other hand, I'm not an experienced user anyway in neither Audiveris or MS.