Sound and Image Engineering in Telecommunication
Optical Music Recognition is the branch of artificial intelligence that develops systems that are able to extract the musical content of an image from a musical score and transcribe it into a format that allows it to be easily processed by a computer. The tendency for the development of these systems is to use automatic learning techniques. These techniques are able to infer the transcription from correct examples of the task, that is, set of pairs (image, transcription). Given the complexity of music, for these techniques to produce satisfactory results it is necessary to use a very large set. To achieve this, a system of automatic generation of tagged data has been developed. This system has two outputs: on the one hand, the ex- pected transcription of the generated score; on the other hand, the score in PDF format.
Technical report in UA repository: