In this paper, we consider an extended version of the traditional note-level transcription task that recognizes onset, offset, and pitch, through including extraction of additional note value to ...