Filter and display benchmark data for unlearning methods
Generate music from text descriptions and optional melodies