Shared spaces for digital music scores – CollabScore
The project is dedicated to the collaborative digitization of music scores currently available only as images in museum collections. The workflow begins with an OMR (Optical Music Recognition) step, producing an incomplete or incorrect score. The second step relies on a crowdsourcing correction effort, in which a community of users will partially improve the scores. The reconciliation step of the different versions will be automatised with specifically developed software, to obtain a reference score. This fulcrum notation will then be used in conjunction with several sources, to enhance the user experience. For instance, listening could be assisted by the synchronized scrolling of the score, and augmented with musicological annotations. The project aims at solving some scientific challenge, first by guiding and controlling an OMR process with musical knowledge, then by elaborating an automated crowdsourcing process. A demonstrator will be delivered, to be used by our user-case partner
Project coordination
Philippe Rigaux (CENTRE D'ETUDES ET DE RECHERCHE EN INFORMATIQUE ET COMMUNICATIONS)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
Partner
BnF Bibliothèque nationale de France
U LILLE UNIVERSITE DE LILLE
IRISA Institut de Recherche en Informatique et Systèmes Aléatoires
FONDATION ROYAUMONT
Antescofo
CEDRIC CENTRE D'ETUDES ET DE RECHERCHE EN INFORMATIQUE ET COMMUNICATIONS
IReMus Institut de recherche en Musicologie
Help of the ANR 570,286 euros
Beginning and duration of the scientific project:
- 42 Months