Morphing and Assessing in a Unified Piano Performance Space
Conditional Generation
MAJEPPA generates expressive performances conditioned on a musical score and a target style.
Each example below shows the same score realized under four different conditions
— from virtuoso concert performance to novice practice.
🎹 Try it yourself:
pick a score, choose a performer / recording condition, and generate live on
our interactive demo.
(First call after the Space wakes takes ~60–90s.)