| • Tore_Danielsson_WMSE | |
| Oct 23 2020, 10:44 AM |
Ask Musikverket or Levande musikarv (or other music institutions) about which music sheets are preferably to work with, for example generate Midi-files or LilyPond. A test can be done if it also can be used as round tripping.
Work with structure, categories and do some test test of OMR for overview of the field.
Notes from discussion here: https://docs.google.com/document/d/1YjRhtE3CvX40KxHtKMOusd1HwnEZ6XfyCiLEwsQYHNM/edit#
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | Jopparn | T266236 Call for data sources to work with in Wikidata | |||
| Resolved | Jopparn | T266323 Work with musical notes on Wikimedia's platforms | |||
| Resolved | • Tore_Danielsson_WMSE | T273437 Make a new recording of music – Alice Tegnér |
Work with the music from Alice Tegnér – https://musikverket.se/svensktvisarkiv/alice-tegner-150-ar/
Added sheet music for Alice Tgenér to Wikimeda Commons: https://commons.wikimedia.org/wiki/Category:Sjung_med_oss,_Mamma!
Plan for new recording
Test the musical sheets in an OMR program. One advice is to use musical sheets from after 1920 – due to the program can better transform to more correct music – this can be tested with musical sheets from Alice Tegnér.
Also start to modelling "Sjung med oss Mamma!" in Wikidata. Every song can be an object in a hierarchy. See T266319
The challenges described are very interesting. Please make sure that this is covered in the final report!
The testing of for example Musescore is not mentioned in the report due to the problems to get any good result with a minor effort. The programs seems to be to rough to be working with and if you get some result it had obvious errors in the playing. This is a threshold and perhaps you have to be a musician (or read and understand sheet music) to fully work with the programs and transcribe from old music sheets to correct machine readable musical sheets.
However, this is an suggestion from our partner, Swedish Performing Arts Agency, to work further with this. Perhaps this can be the apply for next project? This is mentioned in the report and we have had discussions about this scenario.
There are ongoing research in Sweden in this field and perhaps we can be a part or find new project partners as The Royal Swedish Academy of Music for example