SOURCE-LINKED INTELLIGENCE

CANTABILE: Learning Expressive Dynamics for Robotic Piano Performance

arXiv · AI, language, vision and robotics · article · Sep 16, 2026 · UTC

Robotic piano playing has emerged as a standard benchmark for dexterous bimanual manipulation, yet progress on it has been measured almost entirely by note accuracy -- which keys are pressed (pitch) and when (onset) -- leaving the musical dynamics essential for expressive performance neither rewarded nor evaluated. We propose CANTABILE, a dynamics-aware framework for robotic piano performance that (i) closes the score-to-contact loop by conditioning the policy on upcoming velocity goals and mapping each key's angular velocity at onset back to MIDI velocity, (ii) couples a velocity-fidelity rew

Read original source ↗ Open in workspace

recordType

paper

region

Global