A new multimodal model called MusicFormer fuses audio, sheet music, lyrics and regional cultural labels to recognize regional ...