Folding music: modeling transformations between voice-leading spaces