On http://kenai.com/projects/audiveris/ you'll find all related management features (source code, wiki, [mailing list], download area, issue tracking, etc).
Audiveris is an Optical Music Recognition (OMR) module. Starting from the image of a music sheet, it provides high-level logical music information compliant with the MusicXML definition. Other tools such as a Midi Sequencer, or a Composition Editor can then read and update this data.
There are already commercial tools in this area but Audiveris is, to our knowledge, the only open-source OMR tool. It is a cross-platform tool, written in Java, and tested on Windows and Linux. Mac OS would require a volunteer.
Audiveris works with printed music sheets only, the task of recognizing hand-written scores being significantly harder.
Input | |
Input image formats: | All major formats: gif, png, tiff, bmp, etc. |
Input PDF files: | Yes |
Direct scanner input: | No (Lack of open-source twain interface) |
|
|
Processing | |
Support for handwritten scores: | No |
Number of pages per score: | One (Improvement is planned) |
Number of staves per system: | Any (And can vary between systems) |
Number of lines per staff: | 5 |
Support for tablatures: | No |
Number of voices per measure: | Any |
Customizable parameters: | Yes For detailed tuning via user interface |
Integrated learning: | Yes To train on specific symbol shapes |
User plugins: | Yes To allow user extensions |
|
|
Output | |
MusicXML file: | Yes Using MusicXML version 2.0 |
Integrated playback: | Yes At score level and at measure level |
MIDI file: | Yes |
java -jar <install-dir>/dist/audiveris-3.3.jar
Note: The previous Audiveris version required both the installation of Microsoft C/C++ and the setting of TESSDATA_PREFIX variable for Tesseract OCR. None of these is needed anymore, OCR features work out-of-the-box for Windows and Linux.